You are on page 1of 296

Arsuaga / Chosen Species 1405115327-1-pretoc Final Proof page 1 9.6.

2005 9:37am

The Chosen Species

Arsuaga / Chosen Species 1405115327-1-pretoc Final Proof page 2 9.6.2005 9:37am

For Lourdes
For Chon
Arsuaga / Chosen Species 1405115327-1-pretoc Final Proof page 3 9.6.2005 9:37am
Arsuaga / Chosen Species 1405115327-1-pretoc Final Proof page 4 9.6.2005 9:37am

Spanish edition ß 1998 by Juan Luis Arsuaga Ferreras, Ignacio Martı́nez Mendizábal, and
Ediciones Temas de Hoy, Ltd. (T.H.)
Illustrations ß 1998 by Mauricio Antón
English translation ß 2006 by Blackwell Publishing, Ltd.
350 Main Street, Malden, MA 02148-5020, USA
9600 Garsington Road, Oxford OX4 2DF, UK
550 Swanston Street, Carlton, Victoria 3053, Australia
The right of Juan Luis Arsuaga Ferreras and Ignacio Martı́nez Mendizábal to be identified as
the Authors of this Work has been asserted in accordance with the UK Copyright, Designs,
and Patents Act 1988.
All rights reserved. No part of this publication may be reproduced, stored in a retrieval
system, or transmitted, in any form or by any means, electronic, mechanical, photocopying,
recording or otherwise, except as permitted by the UK Copyright, Designs, and Patents Act
1988, without the prior permission of the publisher.
Originally published in 1998 by Temas de Hoy as La especie elegida
English edition published 2006 by Blackwell Publishing
Translated into English by Rachel Gomme
1 2006
Library of Congress Cataloging-in-Publication Data
Arsuaga, Juan Luis de.
[Especie elegida. English]
The chosen species : the long march of human evolution / Juan Luis Arsuaga, Ignacio
Martı́nez ; illustrations by Mauricio Antón ; [translated into English by Rachel Gomme].
p. cm.
‘‘Originally published in 1997 by Temas de Hoy as La especie elegida’’—T.p. verso.
Includes bibliographical references and index.
ISBN-13: 978-1-4051-1532-2 (hard cover : alk. paper)
ISBN-10: 1-4051-1532-7 (hard cover : alk. paper)
ISBN-13: 978-1-4051-1533-9 (pbk. : alk. paper)
ISBN-10: 1-4051-1533-5 (pbk. : alk. paper)
1. Fossil hominids. 2. Human evolution. 3. Primates—Evolution. 4. Anthropology,
Prehistoric. I. Martı́nez, Ignacio. II. Antón, Mauricio. III. Title.
GN282.A69813 2006
A catalogue record for this title is available from the British Library.
Set in 10/12.5pt Galliard
by SPI Publisher Services, Pondicherry, India
Printed and bound in the United Kingdom
by TJ International Ltd, Padstow, Cornwall
The publisher’s policy is to use permanent paper from mills that operate a sustainable forestry
policy, and which has been manufactured from pulp processed using acid-free and elementary
chlorine-free practices. Furthermore, the publisher ensures that the text paper and cover
board used have met acceptable environmental accreditation standards.
For further information on
Blackwell Publishing, visit our website:
Arsuaga / Chosen Species 1405115327-1-toc Final Proof page 5 8.6.2005 5:18pm


Acknowledgments xi

Introduction: Prehistory 1
Little Lucy 1
Intrepid Paleontologists 3

Part I Children of Africa 7

1 Basic Principles of Evolutionary Theory 9

The Inheritance of Acquired Characteristics 9
Natural Selection 10
The Staircase of Progress 13

2 We the Primates 17
The Ecological Definition and Diversity of Primates 17
The Classification of Primates 20
Hominoids, Apes of Our Own Branch 22
History of the Primates 27

3 Climate and Evolution 35

The Origin of Species 35
Climate Changes over the Last Few Million Years 36
Paleotemperature Scales 38
Factors in Climate Change 39

Arsuaga / Chosen Species 1405115327-1-toc Final Proof page 6 8.6.2005 5:18pm


Milankovitch Cycles 40
A Climatic Model for Equatorial Africa 43
The Controversial Gas 45
The End of Paradise 46

4 The Origin of Humanity 49

Molecular Clocks 49
The First Fossil Hominids 51
Change of Habitat 53
East Side Story 57
Dating Fossils 59
The Taung Child 61
Distinguishing Marks 62

5 The Bipedal Chimpanzee 66

The Great Step 66
The Laetoli Footprints 74
The Mystery of Mysteries of Human Evolution 76
Portrait of the Entire Body of an Australopithecine 78

6 Paranthropus – Hominids of the Open Plains 87

The Emergence and Distribution of Paranthropus 87
The Specialist 94

7 A New Kind of Hominid 97

The First Humans 97
The Stone Cutters 98
The Diversification of Homo 102
Ready for the Great Leap 105
Family Relationships 107
The Science of Relationships 110
The Hominid Tree 112

8 The Evolution of the Brain 116

The Organ of Intelligence 116
Large and Small Brains 117
World Champions of Encephalization 120
Weighing Ghosts 120
The Brain Size of Fossil Hominids 123
Surface Area of the Brain 124
The Size of the Intellect 127

Arsuaga / Chosen Species 1405115327-1-toc Final Proof page 7 8.6.2005 5:18pm


9 Teeth, Guts, Hands, Brain 129

Types of Diet 129
Carnivorous and Herbivorous Mammals 132
The Teeth of Primates 134
The Teeth of the First Hominids 136
Size of the Molars and Shape of the Hand 138
Guts and Brain 142

10 Development 145
The Rhythm of the Molars 145
Birth and the Newborn 146
Childhood and Adolescence 152

11 Social Intelligence 155

The Unexciting Sex Life of the Female Orangutan 155
Behavior as Adaptation 156
Comparative Sociobiology of Hominoids 158
Natural Selection and Sexual Selection 160
Bipedal and Monogamous from the Beginning? 162
Brain Size and Size of Social Group 169

Summary 171

Part II A New Home 177

12 New Locations for Human Evolution 179

Homo erectus and the Settlement of Asia 179
The First Europeans 182
Gran Dolina and the First Europeans 184
Prehistoric Cannibalism 186
Homo antecessor 187
Human Evolution in Europe during the Middle
Pleistocene 190
The Pit of Bones 192

13 The Neanderthals 197

The Way They Were 197
Life and Death among the Neanderthals 205
The Beginning and End of the Neanderthals 208

Arsuaga / Chosen Species 1405115327-1-toc Final Proof page 8 8.6.2005 5:18pm


14 The Origins of Modern Humanity: The Fossil Evidence 212

Neanderthals and Modern Humans 212
Two Intelligent Human Species 216
The Levant: A Crossroads 219
To the Easternmost Edge of Asia 222
The African Origin of Homo sapiens 223

15 The Origins of Modern Humanity: The Genetic

Evidence 226
A Brilliant Idea 226
The Molecules of Inheritance 228
Black Eve 229
An Adam for Eve 231
The Other Chromosomes 232
Pleistocene Park 233
Fossils and Molecules 235
Standards of Beauty 237

16 The Origins of Human Language 239

King Solomon’s Ring 239
Language and Brain 241
The Choking Primate 243
The Production of Speech 244
The Fossils Speak 247
Group Selection and the Extinction of the Neanderthals 249
The Crooked Lines of Natural Selection 253

17 The Meaning of Evolution 255

The Action Replay of Life 255
Organization and Chaos 261

Epilogue 264
The Never-ending Story 264

Bibliography 268

Index 275

Arsuaga / Chosen Species 1405115327-1-posttoc Final Proof page 9 4.6.2005 2:31pm

For now consider, like the zoologists and the anatomists,

that man has more of the ape than of the angel, and that
he has few grounds for vanity or conceit.
Santiago Ramón y Cajal
Arsuaga / Chosen Species 1405115327-1-posttoc Final Proof page 10 4.6.2005 2:31pm
Arsuaga / Chosen Species 1405115327-1-posttoc Final Proof page 11 4.6.2005 2:31pm


The authors are indebted to many people whose contributions helped to

improve this book. First and foremost to Jesús Arsuaga, who devoted
much time to investigating and summarizing what is known today about
climate changes and their influence on human evolution. The majority of
the chapter on climate is his work. Manuel Martı́n-Loeches made very
interesting suggestions on the subject of the brain. Elena Benavente and
Jesús Pérez-Gil reviewed the genetic evidence. We, and we hope the book
also, benefited from our ‘‘philosophical’’ conversations with Fernando
Palacios. And from our co-researchers José Miguel Carretero, Nuria
Garcı́a, Ana Gracia, and Carlos Lorenzo, we consistently received gener-
ous support and valuable ideas.
Arsuaga / Chosen Species 1405115327-1-posttoc Final Proof page 12 4.6.2005 2:31pm
Arsuaga / Chosen Species 1405115327-4-intr Final Proof page 1 4.6.2005 2:47pm

Introduction: Prehistory

The main conclusion arrived at in this work, namely, that man is

descended from some lowly organized form, will, I regret to think, be
highly distasteful to many.
Charles Darwin, The Descent of Man

Little Lucy

Little Lucy trudged her way laboriously through the African savanna. As a
result of continuous effort over generations, she walked on two legs rather
than four like her ancestors. With her child weighing heavy in her arms,
she felt herself weakening as she drew nearer to the clump of thorny
acacias which could be glimpsed on the horizon, under the torrid tropical
sun. Small as she was – little more than a meter in height, and weighing
less than 30 kilograms – it was only by her wits that she had evaded
powerful predators. She had no stone tools. A million years had passed
since her ancestors, the first hominids, had decided to leave the protection
of the forest and move out into the savanna which continued to expand,
as a result of the great climate change then taking place. Her relatives,
the ancestors of chimpanzees and gorillas, had preferred the safety of the
forest and would remain there forever. For destiny belonged to the
bold, those who defied the danger of open environments. One day these
bolder beings would evolve, their brains and their intelligence would
develop, they would manufacture all kinds of tools, they would discover
fire and would banish forever the lion, the leopard, and the hyena.

Arsuaga / Chosen Species 1405115327-4-intr Final Proof page 2 4.6.2005 2:47pm

Introduction: Prehistory

Eventually, they would conquer the world. All of this depended on Lucy
and her baby surviving and joining the little group of australopithecines
waiting in the stand of trees, who represented the future of humanity.
Lucy literally carried our future on her two legs.
But no, dear reader, this is not one of those books; we have not allowed
our imagination to run away with us. This is not a fanciful tale about
human evolution. Its aim is not to keep you in suspense, holding your
breath over the vicissitudes of our ancestors. After all, the ending of the
story of human evolution is well known. Lucy survived, if you like to think
of it those terms. Ultimately we are all descended from many Lucys.
Misia Landau has drawn attention to the narrative structure of histories
of human evolution, the rhetoric which surrounds them and its similarity
to mythological and religious literature. This is evident in the biblical
references in some of the names given to scientific hypotheses about our
origin, such as the Black Eve hypothesis, the Noah’s Ark hypothesis, or
even the title of this book. In fact in other cases the only aim of such
names is to catch the reader’s attention with familiar-sounding titles, like
the Out of Africa hypothesis, or the East Side Story hypothesis. But the
really important issue is not the name given to these hypotheses, but that
they can be judged against the facts, modified and even rejected if they are
not compatible with them: this is what makes them scientific hypotheses
rather than simply opinions or fantasies. Only dogma is immutable.
For however her story is told, Lucy is much more than a myth. Lucy is
real. The man who discovered her, Donald Johanson, gave this name in
1974 to an extremely well-preserved skeleton of a female hominid who
lived in what is now known as Ethiopia some three million years ago. And
she really was small. She no longer lived in the enclosed, humid forests,
but rather in more open, arid spaces: although this environment increased
the risk of falling victim to predators, it offered new plant resources for
food. Lucy could not speak as we do; her brain was not much bigger than
that of a chimpanzee, and she had no stone tools, but she was bipedal. All
the other elements of our tale – the split between the lines leading to
chimpanzees and to our species, when and where it took place, the climate
change which occurred and the reduction in the tropical forest which
resulted from it – all of this is based on scientific data. Our only artistic
license is in relating the life-story of a particular hominid individual.
Nevertheless, there are elements in the structure of this story which
remain hidden from view, but which have profound implications for
correct understanding not only of human evolution, but of evolution
in general. It is therefore worth pausing a moment to analyze these

Arsuaga / Chosen Species 1405115327-4-intr Final Proof page 3 4.6.2005 2:47pm

Introduction: Prehistory

First, the tale is evolutionist: it accepts that our species originated

through evolution from other species, thus forming a long, continuous
chain through time. However, the structure of the story as we presented it
here is not Darwinist, because it contains a subtle acknowledgment of the
active role of living organisms in their own evolution, setting them as
protagonists confronted by a changing environment. What is more, in the
first paragraph of the story the development of an erect walking position is
attributed to continuous effort and exercise. Darwin, on the contrary,
believed that organisms are passive subjects in evolution: they constitute
the raw material which natural selection molds, giving form to different
and changing species over time, and the activities undertaken by individ-
uals during their lifetime have not the slightest effect on the anatomical
structures and organs which they will pass on to their offspring.
Finally, and most importantly of all, the story with which we open this
book does not present evolution as a process directed by either internal or
external forces which guide it, according to a preestablished plan or
design, towards its culmination in the human being. Lucy could easily
die and her lineage be lost; the entire species could have become extinct,
and we would not be here today. In other words, the implication of this
tale is that we are not the inevitable consequence of the evolutionary
process; we were at the mercy of fate.
The majority of the people with whom we have discussed human
evolution outside of our professional sphere had no problem with ac-
knowledging the evolutionary origin of our species, but they were con-
vinced nevertheless that we are the ‘‘most evolved’’ species, the
culmination of the entire evolutionary process – in short, the ‘‘chosen
species.’’ Since this appears to be a widely held conviction, and since in
order to discuss it we need to understand how evolution occurs, what it
consists of, and where it is headed (if indeed it is headed anywhere), we
shall devote the first and the last chapters of this book to this crucial

Intrepid Paleontologists

This is also not a book about daring paleoanthropologists and their

adventures discovering human fossils, although we could tell some of
these tales from first-hand experience.
Scientific research is always an intellectual adventure, one which poses
challenges, attempts to reach new horizons of knowledge, and must

Arsuaga / Chosen Species 1405115327-4-intr Final Proof page 4 4.6.2005 2:47pm

Introduction: Prehistory

overcome numerous obstacles with large doses of ingenuity and effort.

But paleontology is doubly an adventure, because the objects of study
must be sought out in the field itself, in nature. When scientists speak of a
discovery, they are generally referring to the discovery of some law or
property, or the solution to a complex problem which may perhaps be
expressed in terms of a formula. In paleoanthropology, in addition to this
type of discovery, a new find may, in exceptional cases, take the most solid
and material of forms. It might consist of the fossilized remains of one of
our remote ancestors. Paleontologists are the only scientists who are able
to travel far back in time and, in the case of paleoanthropology (the study
of ancient humans), transport us to any given moment in the history of
our origins. We hope that this book will communicate to the reader our
passionate enthusiasm for the search for our ancestors, without the need
to describe our emotions at those unforgettable moments of discovery of
fossil humans – emotions which we shared with our companions in toil
and which, to tell the truth, we could not express in words.
Every time we gave a lecture we could sense the interest which the
subject of human evolution awakens in the most varied of audiences. But
at the end of the discussions those present were too shy to formulate the
questions which occurred to them, because they seemed too basic, un-
worthy of being put to a professional paleontologist. People do not realize
that the questions everyone wonders about are the same as those the
scientist tries to solve, and that they are often the most difficult to answer.
How do we know the age of fossils? Where and when did we emerge?
Have we been ‘‘murderous apes’’ from the beginning of our history?
What came first, a being that walked on two legs or one that was intelli-
gent? Were our ancestors monogamous? Why does childbirth hurt? How
long did childhood last among primitive hominids? What did they eat?
How tall were they? When did humans first begin to speak? Are we the
hominid species with the biggest brain? This book is designed to respond
to these questions. But in order to answer them, they need to be posed in
the right way, and placed in the context of human evolution.
To some extent, the work of the paleoanthropologist resembles detective
work. Like the detective, the paleoanthropologist arrives at the scene of
the ‘‘crime’’ after it has already happened. On the basis of indirect infor-
mation, he must reconstruct the sequence of events and – the more difficult
part – find logical explanations which enable us to understand what has
happened; both the detective and the paleoanthropologist must give an
account of the ‘‘how’’ and the ‘‘why’’ of the events that took place.
Good detective novels provide the reader with all the clues and the
detective’s deductions to elucidate the case. It is very irritating to read to

Arsuaga / Chosen Species 1405115327-4-intr Final Proof page 5 4.6.2005 2:47pm

Introduction: Prehistory

the end of the novel only to find that the solution depends on evidence
which only the detective knew about and which has been concealed from
the reader until that point. But what is really unforgivable in a detective
novel is if the solution to the case is not explained in a satisfactory way,
since the most interesting thing is not ‘‘whodunit,’’ but how it was
worked out. This is because the detective novel appeals to the reader’s
intelligence. But if the reader is to have all the information at his disposal
and then appreciate the detective’s shrewd work, he needs to be present at
the interviews with all the witnesses, to observe the scene of the crime at
his leisure, to investigate suspects’ past history, study the results of labora-
tory analysis, and take time to reflect in order to try and fit all the pieces of
the puzzle together.
Well, this book also appeals to the intelligence of our readers, and we
have therefore tried to ensure that you will find in its pages all the facts and
arguments on which our conclusions are based. To this end, the book
follows the chronological axis of the evolution of hominids, and tackles
the key questions at different points. As in a detective novel, the reader can
skip straight to the end to find the solution, but this will mean missing
most of the plot; moreover, the file remains ‘‘open’’ on many of the
questions of human evolution.
Two quotations cited in Fernando Trueba’s Diccionario del cine [Dic-
tionary of Cinema] influenced the conception of this book, we hope to
the good. One of these will be discussed in the epilogue. The other is
apposite here, because it was a great solace when we were attempting to
make some of the really very complicated scientific problems easily com-
prehensible for a general audience. Under the keyword ‘‘simplicity’’ in
Trueba’s book, he quotes the following phrase from Albert Einstein:
‘‘Everything must be made as simple as possible, but no simpler.’’

Arsuaga / Chosen Species 1405115327-4-intr Final Proof page 6 4.6.2005 2:47pm
Arsuaga / Chosen Species 1405115327-4-001 Final Proof page 7 4.6.2005 2:32pm

Children of Africa
Arsuaga / Chosen Species 1405115327-4-001 Final Proof page 8 4.6.2005 2:32pm
Arsuaga / Chosen Species 1405115327-4-001 Final Proof page 9 4.6.2005 2:32pm

Basic Principles of
Evolutionary Theory

There seems to be no more design in the variability of organic beings

and in the action of natural selection, than in the course which the
wind blows.
Charles Darwin, Autobiography

The Inheritance of Acquired Characteristics

It would appear logical that the evolution of species is directly related to

the life habits of the individuals which make up those species. If our
ancestors lived in and moved through trees, over the course of their
individual lives they would naturally learn to jump and climb. We might
assume that this would result in changes in their physical constitution
which would be inherited by their offspring, who would in their turn
refine the modifications and pass them on, with improvements, to the
succeeding generation. If at some point a group of apes began to come
down from trees and adopted the habit of walking on the ground on two
feet, the exercise of this activity would make things easier for their des-
cendants. The latter, by continuing in this new form of locomotion,
would gradually, through use and over generations, modify the anatom-
ical structures necessary for walking upright.
This was how Jean-Baptiste de Lamarck (1744–1829) understood
evolution at the beginning of the nineteenth century. His theory was
based on the principle that the transformations occurring in individuals
over their lifetime, through the use or lack of use of organs and structures,

Arsuaga / Chosen Species 1405115327-4-001 Final Proof page 10 4.6.2005 2:32pm

Basic Principles of Evolutionary Theory

would be transmitted to their offspring. In his best-known example, he

suggested that giraffes had acquired their long necks by stretching them,
over generations, to reach the leaves of trees.
Although this explanation seems quite reasonable, unfortunately for
Lamarck the natural world is not governed by human logic. What we
have discovered of the laws of heredity, from Gregor Mendel (1822–84)
up to now, leads us to reject Lamarckian theory. Whatever we do, we
cannot modify the genes our children will inherit. However much we
swim during our own lifetime, they will have to start from square one. The
laws of biological inheritance are not like human laws.

Natural Selection

Years later, an alternative to Lamarck from within the evolutionist camp

was put forward in the work of Charles Darwin (1809–82) and Alfred
Russell Wallace (1823–1913). These scientists took the view that individ-
uals had no active role in evolution. The resources of the environment
being limited, only a few of those that are born will succeed in reprodu-
cing. Given that all the individuals in a given species are genetically
different (except for monozygotic [identical] twins, developed from a
single fertilized ovum), the competition which ensues will see some at
an advantage and others at a disadvantage because of their genes: this is
how selection occurs.
At the end of his life when, in 1876, he was writing a brief autobiog-
raphy for his children, Darwin continued to be amazed at how innumer-
able organisms of all types were so wonderfully adapted to their life habits
(what today we would call their ecological niche). By way of example he
cited the woodpecker’s adaptation to enable it to climb trees; or seeds
which have developed parachute-like down to help them to disperse on
the wind, or hooks which allow them to attach themselves to animals’ fur.
He explained this in terms not of the will of the organisms themselves, but
of natural selection which, he suggested, preserves favorable variations
and destroys unfavorable ones.
Although Darwin discovered early on that the key to evolution lay in a
selection similar to that which had been practiced on domestic animals
and plants since Neolithic times, it was not until October 3, 1838 that he
came to understand how this principle could be applied to organisms
living in the wild. On that date Darwin read an essay by the economist
and demographer Thomas Robert Malthus (1776–1834), which stated

Arsuaga / Chosen Species 1405115327-4-001 Final Proof page 11 4.6.2005 2:32pm

Basic Principles of Evolutionary Theory

that if they were not checked, human populations would tend to grow in
geometric progression, outstripping the increase in resources.
There is a kind of sentimental ecological thinking which has become
popular among city-dwellers, and which hampers correct understanding
of natural selection. Many people believe that animals, when they are not
persecuted by humans, lead a pleasant, risk-free life in nature. If this were
the case, why would living beings have to adapt to become more efficient?
On the contrary, simple arithmetic shows us that, in the natural world,
the life of animals is far from smooth. In demographically stable popula-
tions – in other words, populations which are not growing, which includes
all populations in normal circumstances over the long term, although
there may be short-term fluctuations – each reproducing pair is replaced
by another two individuals in the succeeding generation. Nevertheless, in
favorable conditions, a zebra living on the African plains gives birth to a
foal each year from the age of 4, over a period of 15 or more years; a
gazelle gives birth to one kid every 6 months from the age of 2. It is
obvious that the majority of those born will not reach adulthood or
reproduce. The predators fare no better: lions begin to reproduce at the
age of 4 (they can easily live up to 15 years in the wild), and have 2 or 3
cubs every 20 or 30 months; leopards living in the wild reproduce from
the age of 2 to the age of 12, producing from 1 to 3 cubs at intervals of
about 25 months.
The same reasoning applies to primates, and, over the period of their
evolution, to humans as well, although our situation has changed in
recent times, since the infant mortality rate has fallen so much that
without birth control we see an explosion of the population. The propor-
tion of fertilized eggs which do not become reproducing adults is almost
one hundred percent in the majority of aquatic vertebrates (amphibians,
fish), and in almost all invertebrates. The upshot of these simple figures is
that individuals in the different species are constantly under the threat of
death, and that consequently small genetic advantages may be crucial to
reaching adulthood and reproducing, or to continuing to reproduce. This
is what Darwin meant when he spoke of the struggle for existence, which
is not necessarily a bloody battle: plants and herbivores also compete with
one another.
Unlike artificial selection, performed by humans with animals and
plants and gradually enhancing specific characteristics in order to increase
productivity, natural selection pursues no particular goal. Moreover, no
genetic variant is better than another in an absolute sense; everything
depends on the environmental circumstances. What is favorable at one
moment may be unfavorable at another. In addition, through the

Arsuaga / Chosen Species 1405115327-4-001 Final Proof page 12 4.6.2005 2:32pm

Basic Principles of Evolutionary Theory

phenomenon known as mutation, from time to time individuals with new

variations are born, but the habits and needs of individuals do not in any
way determine the direction the mutations will take. Nevertheless, these
mutations represent an inexhaustible source of new variations on which
natural selection can act, modifying species over time and pushing for-
ward their evolution. Mutations do not in and of themselves produce new
species; rather they increase the variation of existing species.
Chance also plays an important part in evolution, for example when a
few individuals fortuitously (by pure good luck) survive an ecological
catastrophe which decimates their species, or when a few individuals are
passively transported by natural forces (wind, rivers, or sea currents) to
found a new population. The characteristics of these randomly selected
individuals might not be the most common in the original population, but
they will nevertheless form the starting point for subsequent evolution.
Sometimes a catastrophe on a larger scale may eliminate one or more
perfectly well-adapted species at a stroke, as we shall see below.
In general terms, this is the basic theory commonly accepted by the
scientific world since the 1940s. It is known as neo-Darwinism, because it
integrates, in a modern synthesis, the ideas of Darwin and advances in
genetics and other areas of biology, including the study of fossils, or
Within the evolutionist camp there are those who contest this vision of
a gradual evolution, proceeding by small steps like those patiently taken in
the artificial selection of domestic animal breeds. Authors such as Stephen
Jay Gould and Niles Eldredge believe that evolution advances in great
strides, or even leaps. In other words, the great evolutionary innovations,
the appearance of the large groups of organisms such as birds or verte-
brates, is due not to the gradual accumulation of small changes, but to
radical transformations.
In fact, living organisms are mechanisms so complex, and at the same
time so perfectly well tuned, that it is difficult to understand how mutants
radically different from their progenitors and yet capable of surviving –
what Richard Goldschmidt (1878–1958) called ‘‘hopeful monsters’’ – can
appear. Various explanations of the viability of these ‘‘monsters with a
future’’ have been put forward. These include changes in the develop-
mental process which, acting on both the pathways and rhythms of
development, would result in adults surprisingly different from their
parents – for example, having some exaggerated characteristics, or con-
versely, appearing infantile in certain aspects.
On the other hand, it is also difficult to understand how natural
selection can detect minute changes in order to favor them. The time

Arsuaga / Chosen Species 1405115327-4-001 Final Proof page 13 4.6.2005 2:32pm

Basic Principles of Evolutionary Theory

factor has been invoked to support the model of slow evolution, suggest-
ing that infinitesimal variations would give those who carried them the
tiniest of advantages which would only become dominant over many
generations. And there is no shortage of time in paleontology. After all,
life has existed on earth for more than 3,500 million years. It is around
this tension between two extremes, gradual evolution or evolution by
leaps, that current debate within evolutionary theory is focused.

The Staircase of Progress

Let us broaden our focus, moving from evolution at species level, or

microevolution (which is measured in hundreds of thousands of years),
to look at evolution in its wider context, or macroevolution, which in-
volves whole groups of organisms comprising many species, and operates
on a time scale of millions of years. If we look back at the course of our
own evolution, do we not see a tendency to more and more complex and
intelligent forms, culminating in our species? Are we not the predictable
result of evolution? And is our species not, as is commonly held, the most
evolved of all? And now that we have arrived at this point, is human
evolution complete or will it continue toward even more intelligent,
even more perfect forms?
This theory of evolution as a staircase of progress which leads to the
species Homo sapiens is deep-rooted in our society, and no less so in
scientific and academic circles, albeit subconsciously. For many years one
of the authors taught a university course in human paleontology, and
readers may already have guessed where this course was timetabled in
the undergraduate program: in the last half of the last semester of the final
year. Paleontology books which discuss evolution also place the evolution
of our species in the last chapter, after that of unicellular organisms and, in
strict order, invertebrates, amphibians, reptiles, birds, and the other mam-
mals. Under these circumstances it is difficult for anyone to escape the
idea that evolution ends with us, perhaps for ever. Even terms such as
lower and higher vertebrates, or lower and higher primates, are still used
(naturally, in both cases we are classified among the higher categories).
We doubt whether there is any teaching program or book which begins
with the first forms of life on the planet and ends with sea urchins or
insects (plants are always relegated to the margins), with mammals, pri-
mates, and humans lost among the intervening chapters or lectures. Does
this mean that vertebrates are ‘‘better’’ than invertebrates, that mammals

Arsuaga / Chosen Species 1405115327-4-001 Final Proof page 14 4.6.2005 2:32pm

Basic Principles of Evolutionary Theory

are ‘‘better’’ than reptiles, that primates are ‘‘better’’ than other mam-
mals, and finally, that humans are ‘‘better’’ than chimpanzees and gorillas?
According to Darwin, evolution has no goal, it follows no preconceived
design; it is simply opportunist, and is not directed toward any idea of
perfection. To put it another way, all species (including our own) are
equally perfect, each one marvelously adapted to its life habits through
natural selection. In other words, unlike artificial selection which the
farmer or stockbreeder carries out with a particular aim, natural selection
has no objective. Although in common parlance (as well as in politics and
business) the word ‘‘evolution’’ signifies change for the better, in Dar-
winian terms ‘‘evolution’’ means simply change, nothing more.
Among the mammals, humans are distinguished as tailless, bipedal
primates with a large brain, but otherwise we have few original features.
We still retain five digits on our hands and feet, while horses support
themselves on the third phalange of their single digit. We show nothing
like the transformations which bats or dolphins underwent, to evolve from
their quadruped ancestors. Are we more evolved, in the sense of being
more changed, than they are? Granted, a geranium cannot write a book –
that is one of our specialties – but with the aid of light it can synthesize
organic matter from mineral salts, water, and carbon dioxide; there is no
doubt that the geranium has a well-equipped laboratory, and it is difficult
to see it as an ‘‘inferior’’ being.
But those who prefer to see evolution as an arrow which has been
pointing toward us since the beginning will have to answer the question
of what obscure internal forces could be driving evolution in the right
direction, independent of what happens around it. Or are we really
dealing with forces outside of the natural world? In this case we move
outside of the sphere of science, which is the terrain of this book and its
authors. The object of science is to explain natural phenomena, like the
existence of our species (and all the others), in terms of natural causes.
So, to return to the terrain of science, Lamarck believed in the idea of
progress in evolution. However, the mechanism he proposed as the motor
driving evolution forward was adaptive (like that of Darwin), and did not
take any particular direction (even to ‘‘divert’’ organisms along ‘‘aber-
rant’’ blind alleys). Confronted with this paradox, Lamarck resolved it by
adding to his theory of use and lack of use the idea that all living forms
‘‘tended’’ gradually and inevitably toward ever higher (in other words,
more complex) levels of organization. Lamarck never explained the cause
of this tendency toward perfection, but later authors attributed it to ‘‘vital
impulses,’’ and therefore became known as vitalists, or finalists, because
they believed that evolution had directionality.

Arsuaga / Chosen Species 1405115327-4-001 Final Proof page 15 4.6.2005 2:32pm

Basic Principles of Evolutionary Theory

For those who believe that the Story of Life reflects a program unfold-
ing over time, evolution is in some ways comparable to the process of
development which leads from the embryo to the adult, following pre-
established laws (natural, but which we do not as yet understand very
well). Clearly, those who wish to find significance, meaning, or an inten-
tion in the Story of Life will always have recourse to mysterious internal
forces, still to be discovered or unknowable.
But if it is the case that invisible threads have, from the beginning of
time, guided evolution in a linear and orderly manner to result in us, why
do we find ourselves among such a diversity of living forms? As we shall
see, we are not descended from chimpanzees, but we have a common
ancestor with them. Chimpanzees are our brothers, not our parents. Nor
are we descended from any organism like a sea urchin. Nevertheless, the
phylum of echinoderms, to which sea urchins belong, and that of the
chordates, which includes our species, share a very remote ancestor which
was neither a sea urchin nor a human being. Living species are not ordered
in sequence. What we have here is not a staircase to nowhere, but a tree
with many, many branches and with no trunk or central axis. Evolution is
not linear, but divergent.
Despite all of this some authors, although aware that, whether looking
at the past or the present, evolution does not appear to have moved in one
single direction, still express the view that life followed various tendencies,
and that ours is the tendency of increasing intelligence. However, they
never explain how these tendencies arise: they appear to obey mysterious
impulses which have nothing to do with the adaptation of organisms, but
rather act of their own accord. The definition of primates still frequently
includes the term ‘‘tendency toward cerebral expansion,’’ as if a tendency
could serve in and of itself to characterize an entire group including fossil
and living species, which would thus become a ‘‘unit of evolutionary
destiny.’’ Naturally, contemporary primates that do not exhibit such
cerebral expansion are held simply to represent relics of the past, or ‘‘living
The authors remember a time when groups of primates or hominids
which were not in a direct evolutionary line with our species were de-
scribed as deviant forms; these ideas can still readily be found in serious
texts. If such forms became extinct, as a just punishment for their rebel-
lion, they were tarred as abortive or even aberrant forms, described as
trials or failed experiments (carried out by whom?), as blind alleys, or in
other terms which made it clear that it was not a good idea to diverge
from the path marked out by evolution.

Arsuaga / Chosen Species 1405115327-4-001 Final Proof page 16 4.6.2005 2:32pm

Basic Principles of Evolutionary Theory

But let us leave, for now, the debate between vitalism and materialism,
or finalism and Darwinism. Over the chapters of this book we will review
what we know today of human evolution through its different stages and
circumstances. The last chapter will give us the opportunity to discuss, in
the light of past events, the nature of our history. But first, in the next
chapter, we will start to get to know ourselves a little better, locating
ourselves among the diversity of species of living and fossil primates.

Arsuaga / Chosen Species 1405115327-4-002 Final Proof page 17 8.6.2005 5:19pm

We the Primates

Direct perception of the origins of anything is automatically denied to

our eyes as soon as a sufficient depth of the past is interposed.
Pierre Teilhard de Chardin, The Human Zoological Group

The Ecological Definition and Diversity of Primates

If we observe nature objectively we recognize that, despite its enormous

diversity, there is not an infinite number of forms: in fact, all living species
can be grouped within a limited number of biological types. A lion is
simply a large cat, or conversely, a cat is a small lion. It seems that once
evolution has produced a successful new model for an organism, one
which has been well ‘‘tested,’’ as automobile manufacturers would say,
it begins to ‘‘manufacture’’ different variations within the same ‘‘range,’’
sometimes altering little more than the size. A cat represents the size of
lion appropriate for the hunters of small prey ‘‘sector,’’ and thus aspires
to its corresponding ‘‘market share,’’ in competition with other small
Species can thus be grouped in progressively larger categories, such as
Felidae (for all species of the cat type), mammals, vertebrates, and so on.
The larger the category which groups together a given set of species
(felines, mammals, vertebrates, etc.), the further back in time the common
ancestor will be. We belong to the group of primates – the group of
monkeys and apes. It is therefore not correct to say that we are descended
from apes, as if we were no longer apes ourselves. We are still primates just

Arsuaga / Chosen Species 1405115327-4-002 Final Proof page 18 8.6.2005 5:19pm

We the Primates

as much as any of the other approximately 180 living species in the

group. However, we evolved not from any current species of ape, but
from species that are now extinct, many of which are also ancestors of
other modern primates. Having thus situated our species in its natural
context, and accepted that we are not an anomaly, let us consider what
being a primate consists of, since it is something we cannot avoid.
Primates form a very homogeneous set of species in terms of their
ecological requirements. To generalize, they are mammals which live in
humid tropical forests (rainforests) or monsoon-type subtropical forests,
with seasonal rains and dry periods during which some trees lose their
leaves. It was in this warm forest environment that we evolved, and all
primates therefore show various adaptations to life in the trees.
Of course there are exceptions to this definition. We humans form one
such exception although, as we shall see, this has only been the case for a
few million years. The baboons developed in the more or less open
savanna of Africa, where they accompanied us in our evolution and were
probably ecological competitors; at all events, they always seek the pro-
tection of groups of rocks or trees at night. A similar species, the gelada
(Theropithecus gelada), lives on the Ethiopian highlands, far from the
trees. The macaques are a set of species found almost exclusively in Asia,
some of which live in the cold forests of Japan or in the foothills of the
Himalayas, as well as in many other environments. The only African
macaque, the Barbary macaque (Macaca sylvanus), now lives in the Atlas
Mountains of Algeria and Morocco, north of the Sahara. Although the
Barbary macaques in Gibraltar were introduced by humans, this species
flourished in European ecosystems, settling in latitudes as high as England
and Germany before becoming extinct in Europe. Another species which
is well adapted to open environments is the patas monkey (Erythrocebus
patas), which lives in the savanna and only seeks shelter in the trees to
sleep or at times of danger. When running, the patas monkey can reach a
speed of 55 km per hour, the record for primates.
Currently, apart from ourselves, primates are found in the wild only in
Mexico, Central and South America, Africa, and Asia. There are none in
Europe or in Australasia.
Primates are fairly varied in terms of type of diet: some species are
completely herbivorous while others are omnivorous and also eat small
vertebrates and invertebrates like insects; some primates have specialized
in eating insects. Baboons and chimpanzees sometimes even hunt and eat
other mammals, although their diet is otherwise herbivorous.
Having evolved over a long period while living in the trees, modern
primates share a series of unique special features which enable them to

Arsuaga / Chosen Species 1405115327-4-002 Final Proof page 19 8.6.2005 5:19pm

We the Primates

grasp and climb trees, and jump from branch to branch. One such
adaptation is the first toe (the big toe), which is very large and mobile
and can be opposed to the other toes (except in our case). Primates also
have flat nails on all the digits of hands and feet, rather than the claws of
their ancestors. Some primates have claws, sometimes on all the digits
except for the first toe, but it appears that these were originally flat nails
which converted back into claws at a later date.
The teeth of mammals are very important in paleontology for two
reasons: one is that dentition reflects the type of diet, and the other is
that they are the most frequently found fossils – often the only fossils found
for certain groups. A first step toward the study of dentition is counting the
number of teeth an individual has. All living primates descend from an
ancestor who had 36 teeth. However, not all mammals have the same kinds
of teeth – they are specialized for different functions. Four groups of teeth
are distinguished, from front to back: incisors, canines, premolars, and
molars (Figure 2.1). The primate ancestor of all modern primates had, on
each side of the mouth and at both top and bottom, two incisors, one
canine, three premolars, and three molars. Some primates have modified



Figure 2.1 Human jaw, showing the four types of teeth

Arsuaga / Chosen Species 1405115327-4-002 Final Proof page 20 8.6.2005 5:19pm

We the Primates

this formula over the course of their evolution, losing some elements, but
never increasing the number of teeth in any of the four categories.

The Classification of Primates

Scientists are sometimes thought to study very subtle, obscure character-

istics, using microscopes or other advanced instruments to establish the
classification of organisms. However, the main division of primates into
two large groups known as strepsirrhine and haplorrhine primates (Strep-
sirrhinae and Haplorrhinae) is based on nothing more complex than the
form of the nose and the upper lip (Figure 2.2).
In strepsirrhine primates, as in other mammals, the external nasal
orifices, or nostrils, are surrounded by a hairless, moist area of skin
known as the snout, which continues downward in a lip divided into two
parts along its mid-line, through which it is attached to the gums by a
membrane. To understand this morphology better, the reader simply

Figure 2.2 Strepsirrhine and haplorrhine primates. Left : a ring-tailed lemur

(Lemur catta), a representative of the Strepsirrhinae; right : a pygmy chimpanzee
or bonobo (Pan paniscus), a representative of the Haplorrhinae. Strepsirrhine
primates have hairless, moist skin around the nasal openings, and a divided
upper lip attached to the gums along the mid-line. Haplorrhines have the same
type of skin around the nasal orifices as on the rest of the face, and the upper lip is
undivided and mobile

Arsuaga / Chosen Species 1405115327-4-002 Final Proof page 21 8.6.2005 5:19pm

We the Primates

needs to look at the nearest cat or dog. This arrangement of the nose
severely limits the facial expression of emotions. Haplorrhinae, including
ourselves, do not have this hairless skin around the nostrils, and the upper
lip is undivided and mobile. The continuous upper lip facilitates greater
facial expressivity, a well-known characteristic of haplorrhine primates.
The Strepsirrhinae include lemurs, indris, and the aye-aye (lemurs in the
broader sense), which evolved and diversified in isolation on the island of
Madagascar (off the east coast of Africa). Some species are nocturnal and
others diurnal. Unfortunately, the recent arrival of humans has resulted in
the degradation of their forest paradise and the disappearance of many
The Strepsirrhinae also include the Asian lorises, and the African bush
babies and pottos (the Lorisidae in the broader sense), all of which are
The Haplorrhinae are divided into three groups. One of these is the
tarsiers (Tarsiiformes), small nocturnal primates found in the Philippines,
Borneo, Sumatra, and other Southeast Asian islands, with enormous eyes,
very long tails, and very long back legs, well adapted to jumping. The
other groups of Haplorrhinae are the catarrhines (Catarrhinae), which
include our own species, and the American monkeys, known as platyrrhine
primates (Platyrrhinae).
The catarrhines and platyrrhines are usually grouped together in a
common category used more or less informally, that of anthropoid apes
(technically, Anthropoidea) – from here on we will use the term anthro-
poid. Both groups are diurnal, with the exception of the South American
owl monkey, Aotus trivirgatus, which appears to have become nocturnal,
although it is descended from diurnal ancestors.
Another feature of anthropoids is the completely frontal position of the
eyes, allowing for a wide field of stereoscopic or three-dimensional vision,
which requires that the fields of vision of the two eyes overlap. This type of
vision makes it possible to calculate the distance to objects, whether tree
branches or prey, very precisely. Anthropoids have a large brain, although
it appears that platyrrhines and catarrhines developed this (through evo-
lution) separately. The olfactory lobes of their brains are very much
reduced in size. As anthropoids, we perceive the world essentially in
terms of images, not in odors.
Platyrrhines have the same number of teeth as the first primates, except
for the marmosets and tamarins (Callitrichinae), a group which has lost the
last molar. The catarrhines, however, have lost a premolar (Figure 2.1),
although many of our readers will also find that the last molar (the wisdom
teeth) will never emerge. This absence of the third molar in the adult jaw is

Arsuaga / Chosen Species 1405115327-4-002 Final Proof page 22 8.6.2005 5:19pm

We the Primates

an expression of the reduction in the dental and masticatory (chewing)

apparatus which has arisen in Homo sapiens, our species.

Hominoids, Apes of Our Own Branch

Within the catarrhines, our species is classified among the hominoids,

while the so-called Old World monkeys form the subgroup Cercopithe-
cidae, which includes the macaques, baboons, mandrills, guenons, colo-
bus monkeys, langurs, and others. In addition to ourselves, the hominoid
group also includes a series of primates known as apes. These are, in order
of closeness of relationship to humans, the two species of chimpanzee (our
closest relatives), the gorilla, the orangutan, and the various species of
gibbon, the apes furthest from us in evolutionary terms (Figure 2.3). The
number of species of gibbon varies from five to nine, according to differ-
ent authors, as some recognize as distinct species what others consider to
be simply geographical variants of a single species. If we have problems
classifying living species, imagine the difficulties encountered by the pale-
ontologist working with fossils.
The two species of chimpanzee are the common chimpanzee and the
pygmy chimpanzee, or bonobo (Figure 2.2), which is in fact no smaller
than the common chimpanzee; the two together form a natural group, or
clade, as they are descended from a common ancestor. It should be added,
and this is an important detail, that this common ancestor is exclusive to
these two species – in other words, no other living species descends from
it. The chimpanzees, the gorilla, and ourselves together form another
clade: we also share an exclusive common ancestor. Given that chimpan-
zees and gorillas live in Africa and, as we shall see in this book, we also
come from Africa, it seems reasonable to suggest that the still unknown
common ancestor lived in Africa.
There has been much discussion as to whether chimpanzees or gorillas
are our closest relatives, although genetic studies appear to incline toward
the chimpanzee. In fact, the evolutionary line of the gorilla separated from
the human line a very short time before that of the chimpanzee, so
effectively the three lines separated at virtually the same time. The chim-
panzee line later split to produce the common chimpanzee and the pygmy
chimpanzee, or bonobo.
As can be seen, the system for classification of species created by Karl
von Linné (generally known by the Latin name of Linnaeus, 1707–78) has
a hierarchical structure. We are thus first humans, then hominoids, then

Arsuaga / Chosen Species 1405115327-4-002 Final Proof page 23 8.6.2005 5:19pm

We the Primates

Figure 2.3 Types of hominoid. Both sexes are shown. Top left : a pair of gibbons
(Hylobates lar); top right : orangutans (Pongo pygmaeus); below (left to right): gorillas
(Gorilla gorilla), common chimpanzees (Pan troglodytes), and humans

catarrhines, then anthropoids, then haplorrhines, and finally primates,

which in their turn are mammals, then vertebrates, and so on . . . until
we arrive at the largest category, the animal kingdom.
As a group, hominoids share a set of features inherited from our
common ancestor. Many of these are related to a particular method of
locomotion through the trees which Arthur Keith (1866–1955) termed

Arsuaga / Chosen Species 1405115327-4-002 Final Proof page 24 8.6.2005 5:19pm

We the Primates

brachiation. This form of locomotion consists of traveling by hanging

from branches, with the arms extended, swinging from one arm to the
other as the body turns in the air (Figure 2.3). However, there have been
many variations within the brachiating model, in addition to those cur-
rently found in living species: these represent only a very small part of the
diversity of the group in the past, as we shall see. It would perhaps be more
correct to say that hominoids show adaptations which allow them to hang
from the branches with the trunk erect, rather than traveling over the
branches on all fours or jumping from one to the other, as the arboreal
(tree-living) primates generally do.
In hominoids the thorax is flattened dorsiventrally (from the chest to the
back), rather than being laterally compressed as in the rest of the primates
and in quadruped mammals generally (Figure 2.4). In consequence, our
shoulder blades (or scapulas) are located in a dorsal position, on the back,
rather than laterally, on the sides of the body. The form of the shoulder
blades is different, with a longer vertebral edge (the edge running along-
side the spine). The form of the humerus (upper arm bone) is also different:
the head of this bone (where it articulates with the shoulder blade) is more
rounded, and the diaphysis, or shaft of the bone, twists so that the head of
the bone faces inward rather than toward the back. The expansion or lateral
widening of the thorax also means that the clavicle (collarbone) is longer.
All of these modifications result in a great freedom to move the arms
above the level of the shoulders; combined with the capacity to extend the
arms completely and the mobility of the wrist, this makes brachiation
possible. Since we share these characteristics with other hominoids, we
have to acknowledge that our ancestors must have been brachiators.
Also as a consequence of brachiation, the arms are more fully developed
than the legs. The ratio of the length of the arms to that of the legs among
apes ranges from 147 percent in the siamang (a species of gibbon) to 102
percent in the bonobo. Moreover, the hands lengthen while the length of
the first finger (the thumb) decreases, forming a hook from which to
hang. This modification of the hand makes it difficult for apes to touch
the tip of the index finger and thumb together.
Of course, our ancestors’ subsequent adaptation to bipedal walking
caused some of the characteristics seen in apes to alter, as we shall see in
the chapter on human locomotion (Figure 2.4). In particular, the pelvis and
legs have changed markedly, so that the ratio of the length of the arms to
that of the legs is only 72 percent, while the big toe is no longer opposable
and is aligned with the rest of the toes. Moreover, the thumb has become
longer and the rest of the hand shorter, so that we have recovered the ability
to manipulate small objects which other hominoids have partly lost.

Arsuaga / Chosen Species 1405115327-4-002 Final Proof page 25 8.6.2005 5:19pm

We the Primates

Figure 2.4 Skeletons of primates. Top: colobus monkey (Colobus guereza);

below left : common chimpanzee; below right: human

Arsuaga / Chosen Species 1405115327-4-002 Final Proof page 26 8.6.2005 5:19pm

We the Primates

In hominoids the lumbar region of the trunk is shorter, with a smaller

number of vertebrae (particularly in gorillas and chimpanzees), making us
unable to bend this area as much as other primates can. And as we all
know, we have no tail, although we are not the only primates without one.
Hominoids characteristically hold the trunk erect or vertical, both when
climbing and when swinging through the trees or resting.
Gibbons are sometimes bipedal and travel over the branches, rather
than hanging from them as they habitually would; in this case they use
their long arms to balance. Gibbons rarely come down to the ground,
spending most of their lives in the highest level of the forest. Orangutans
are effectively four-armed rather than four-legged, using their feet, which
are very similar to their hands, to grasp branches or hang from them, and
not traveling much through the trees (Figure 2.3).
When they come down to the ground, orangutans, chimpanzees, and
gorillas move on all four limbs, but the trunk does not become horizontal
as it does in the predominantly quadruped primates: instead it slopes
downward from the shoulders to the hips. When moving over the ground,
which they do very occasionally, orangutans support themselves on
their fists.
In fact, adult gorillas are so heavy that they can barely be considered
brachiators (Figure 2.3). The gorillas’ and chimpanzees’ method of
locomotion on the ground is very particular (Figure 2.5): they support

Figure 2.5 Quadruped locomotion of the chimpanzee. Detail: position of the

bones of the hand

Arsuaga / Chosen Species 1405115327-4-002 Final Proof page 27 8.6.2005 5:19pm

We the Primates

themselves on the soles of their feet and on the dorsal (back) side (not the
knuckles, as is often stated) of the intermediate phalanges of the second to
fifth fingers (the index to the small finger). Quadruped primates walk on
the soles of their feet and the palms of their hands. Gorillas’ and chim-
panzees’ particular mode of travel has also resulted in modifications to the
bones of the arms, to give them more stability when supporting the
weight of the body on either one.

History of the Primates

The first known fossil thought to be that of a primate (though this is not
certain) is a molar from the late Cretaceous, the last period of the Meso-
zoic or Secondary era, when the dinosaurs were still alive. It is around
65 million years old and was found in Montana. It was named Purgatorius
ceratops, as it is contemporary with the well-known three-horned dinosaur
Purgatorius is allocated to the group Plesiadapiformes, which continues
into the following era, the Cenozoic, made up of the Tertiary period
together with the Quaternary, or Pleistocene. Plesiadapiformes (Figure
2.6) are the only primate fossils known from the Paleocene, the first epoch
of the Tertiary (between 65 and 55 million years ago); during this epoch
they split into a number of lines. The last of these became extinct in the
subsequent, Eocene epoch (between 55 and 36 million years ago). The
fact that Plesiadapiformes have been found in both North America and
Europe indicates that the two continents were joined together before they
became definitively separated by the Atlantic Ocean. Throughout the
Cenozoic, Asia and North America were also joined from time to time,
in the region of what is now the Bering Strait. However, Africa was
isolated, like a large island, during most of the Cenozoic, although there
were sometimes land bridges between Eurasia and Africa which allowed
for interchanges of fauna.
There is ongoing debate as to whether the Plesiadapiformes should be
considered true primates or not (for example, they do not appear to have
had flat nails, nor an opposable big toe). Nevertheless, Plesiadapiformes
are the group closest, in evolutionary terms, to living primates, which
themselves form a natural group with an exclusive common ancestor.
Some authors suggest that primates can be divided into two large cat-
egories, Plesiadapiformes, or archaic primates, and others, the euprimates
or ‘‘true’’ primates (see Figure 2.7).

Arsuaga / Chosen Species 1405115327-4-002 Final Proof page 28 8.6.2005 5:19pm

We the Primates

Figure 2.6 Plesiadapis. A Paleocene European plesiadapiform. The skull of

Plesiadapis (top) is compared with that of a red squirrel (below), showing their
similarities – which are due to a similar lifestyle rather than any evolutionary

Euprimates appear in the fossil record in the Eocene (the epoch be-
tween 55 and 36 million years ago), and are well represented in Eurasia
and North America by two large groups, the Adapiformes and the Omo-
myidae, although both groups probably also lived in Africa. The Adapi-
formes show some general similarities with lemurs and lorises, but we
cannot say for certain that they are the direct ancestors of these animals
(Figure 2.8). Many authors consider the Omomyidae to be haplorrhines,
but this classification is also open to doubt. Finally, some middle Eocene
fossils from China and Algeria (about 45 million years old) have been
identified as anthropoids, though not all authors recognize them as such.
The transition from the Eocene to the subsequent epoch, the Oligo-
cene, is well represented in the El Fayum deposits in Egypt (Figure 2.9).
Numerous anthropoid fossils, between 30 and 37 million years old, have
been found here. The anthropoids, let us recall, are the group formed by
the platyrrhines, or New World monkeys, together with the catarrhines,
which in their turn are divided into the Old World monkeys and the
hominoids; the latter comprise apes and humans.

Arsuaga / Chosen Species 1405115327-4-002 Final Proof page 29 8.6.2005 5:19pm

We the Primates

Strepsirrhinae Haplorrhinae

Lemurs and lorises

0 Pleistocene






Years (millions)



Cretaceous Paleocene



Figure 2.7 Diagram of the evolutionary relationships of primates

How platyrrhine primates arrived in South America is a big question,

which has yet to be resolved. At that time South America was an island,
separated from North America by the Caribbean Sea (without the Panama
isthmus), and from Africa by the Atlantic Ocean. Given that no anthro-
poid fossils have been found in North America, it seems unlikely that they
could have come from there. The Atlantic Ocean had not yet expanded to
its present size, and it is possible that at least one pair of primates could
have traveled accidentally from Africa, on a natural ‘‘raft’’ formed by trees
tangled together, if they had encountered favorable currents and winds.
(This is how the ancestors of the lemurs, indris, and aye-aye now found
on Madagascar must have arrived on the island: Madagascar has been
separated from the African continent for more than 100 million years.)
However that may be, the oldest known platyrrhine is Branisella bolivi-
ana, which is around 27 million years old, and it remains a mystery as to
how it arrived in South America. There is another group of mammals that
reached the continent in similar, and similarly unknown, circumstances:

Arsuaga / Chosen Species 1405115327-4-002 Final Proof page 30 8.6.2005 5:19pm

We the Primates

Figure 2.8 Notarctus. A North American Eocene adapiform, with hind limbs
longer than its front limbs and a capacity for jumping similar to that of the modern
lemurs of Madagascar

these are the caviomorph rodents, which include guinea pigs, capybaras,
chinchillas, viscachas, and so on.
During the Miocene, the epoch which succeeded the Oligocene and
lasted from 24 million to 5 million years ago, the first hominoids – the
group to which we and the apes belong – emerged on the African
continent. The oldest fossils, found in East Africa, have been assigned to
the genus Proconsul (Figure 2.10), a type of hominoid which lived before
the line which led to the gibbons split off. The name Proconsul (pre-
Consul) was coined in 1933 in honor of a famous chimpanzee at London
Zoo called Consul, in the mistaken idea that the fossil species was the
direct ancestor of modern chimpanzees.
After Proconsul come Morotopithecus, Afropithecus, and Kenyapithecus.
Fossils of these four species dated between 23 and 14 million years ago
have been found. Proconsul does not appear to have developed the
locomotive adaptations characteristic of modern hominoids. However,
some details of the lumbar anatomy and of the joint between the shoulder
blade and humerus of Morotopithecus, 20.6 million years old, suggest that

Arsuaga / Chosen Species 1405115327-4-002 Final Proof page 31 8.6.2005 5:19pm

We the Primates

Figure 2.9 Oligocene fossil anthropoids from El Fayum. Foreground: Aegyp-

topithecus and its skull; background: Apidium

this genus may offer the oldest fossil evidence of a body organization
similar to that of living apes.
Around 17 million years ago, Africa (at that time joined to the Arabian
peninsula), which had remained quite isolated since the Eocene, drew

Arsuaga / Chosen Species 1405115327-4-002 Final Proof page 32 8.6.2005 5:19pm

We the Primates

Figure 2.10 Proconsul. The first fossil hominoid, which lived in East Africa
during the early Miocene

closer to Eurasia; the Indian subcontinent also became connected to

Eurasia. This allowed hominoids to spread throughout the Old World
and to diversify widely into many genera – each one comprising a number
of species – such as Dryopithecus, Sivapithecus, Lufengpithecus, Ourano-
pithecus, Ankarapithecus, and Gigantopithecus. Fossils of these genera
have been found in Europe, China, Turkey, India, and Pakistan, and are
dated at between 13 and 7 million years old. There seems to be general
agreement that there is a relationship between Sivapithecus and the
orangutans, but the evolutionary position of the other Eurasian homin-
oids is the subject of much debate. Louis de Bonis and George Koufos
maintain that Ouranopithecus, found in Greece (and also known as Grae-
copithecus), is related to the group comprising ourselves, the chimpanzees,
and the gorillas.
In the Can Llobateres deposit, in Catalonia, Spain, Salvador Moyà-Solà
and Meike Köhler recently discovered a large part of a skeleton of

Arsuaga / Chosen Species 1405115327-4-002 Final Proof page 33 8.6.2005 5:19pm

We the Primates

Dryopithecus laietanus. With its long arms and short legs, this fossil gives
us an idea of the way in which these hominoids, of about 9.5 million years
ago, moved around (Figure 2.11). They appear very similar to modern
orangutans, which hang from the branches of trees and move slowly
through them. A well-known but enigmatic long-armed primate known

Figure 2.11 Dryopithecus. Fossil hominoid which lived in Europe during the

Arsuaga / Chosen Species 1405115327-4-002 Final Proof page 34 8.6.2005 5:19pm

We the Primates

as Oreopithecus, which lived during the same period in the swampy forests
of central Italy and Sardinia (which at that time formed a single large
island), could be close, in evolutionary terms, to Dryopithecus. However,
in view of the very specific characteristics of its dentition (type and ar-
rangement of teeth), some authors do not even recognize Oreopithecus as
a hominoid.
The fossil trace of Eurasian hominoids disappears about 7 million years
ago, until forms already very close to the modern orangutan and gibbons
appear; the one notable exception is Gigantopithecus. A species of this
genus (Gigantopithecus blacki) survived in China and Vietnam until only a
few thousand years ago, and therefore must have lived at the same time as
humans. We have three jawbones and more than a thousand individual
teeth from this species. The first fossils of Gigantopithecus blacki to be
found were four molars which the paleontologist Ralph von Koenigswald
(1902–82) purchased in drugstores in Hong Kong and Canton between
1935 and 1939 (in traditional Chinese medicine, fossils are credited with
healing properties). The Gigantopithecus jawbones are larger than those of
gorillas – particularly one of them, believed to be that of a male. These
were probably the largest primates which ever existed.
The front teeth (incisors and canines) were relatively small, while the
premolars and molars were large and wide, with a thick layer of enamel.
The jawbones are also very robust. In these characteristics they are similar
to Paranthropus, a type of hominoid with very strong teeth and jaws,
which we shall describe later. This similarity has led some people to believe
that Gigantopithecus belonged to our evolutionary group, but in fact this
is a case of adaptive convergence – in other words, a resemblance due to
more or less similar chewing activity in two independent lines. The
evolutionary affinities of Gigantopithecus are more likely to be with Siva-
pithecus and other species in the orangutan line.
In view of their size, we can only assume that these enormous apes lived
on the ground, and that they fed on what must have been very abundant
plant resources. The strength of the jawbones and the type of dentition
indicate that they ate a hard, fibrous type of plant which required much
chewing. Some authors believe that this may have been bamboo, eaten by
modern pandas.

Arsuaga / Chosen Species 1405115327-4-003 Final Proof page 35 8.6.2005 5:23pm

Climate and Evolution

The hay appeareth, and the tender grass sheweth itself, and herbs of the
mountains are gathered.
Proverbs 27:25

The Origin of Species

The origin and the disappearance of species are two of the fundamental
problems of evolutionary biology. By selecting animals and plants, man
has managed to produce many different breeds and varieties but not, so
far, any new species. All breeds of a given species can interbreed, produ-
cing fertile offspring which can reproduce again. This is not the case with
different species of animals: although two different species can be bred
together, such as horses and donkeys, their offspring – in this case the
mule – is sterile. Obviously this may be a matter of time, given that we
have only been domesticating animals and plants on a large scale for
10,000 years.
During his famous voyage of five years and two days around the world
in the brig Beagle, Charles Darwin began to get an idea of a basic
mechanism for the development of new species. If a few individuals
remained isolated in a marginal region, they could adapt to the conditions
prevailing there and transform into new species. This would explain the
diversity of chaffinches found in the Galápagos Islands (Ecuador). Each
island had its own distinct species of chaffinch – in some cases more
than one – with different specializations (or, as an ecologist would say,

Arsuaga / Chosen Species 1405115327-4-003 Final Proof page 36 8.6.2005 5:23pm

Climate and Evolution

occupying different niches); all developed from a single species, which had
come from the American continent.
The emergence of new species is not always the result of a seed or a pair
of a particular animal species being passively transported to a remote place
by winds or currents, as occurred with the Galápagos chaffinches or the
platyrrhine monkeys of South America. The area of distribution of a
species may be split by a new geographical barrier, giving rise to new
species on either side of the barrier. And there are also cases where species
are subjected to a change of landscape without having to move or be
isolated by changes in topography. This is the case of the hominoids,
which lost their environment because of a great, planet-wide ecological

Climate Changes over the Last Few Million Years

Climate change is currently a topic of great concern. We fear that human

activity may lead to the earth becoming too warm. It seems to us that it is
hotter than when we were young. We hear that the desert is advancing
and the ice on the mountains and at the poles is melting. In short, the
subject of climate change has become fertile territory for the doomsayers
of the end of the second millennium. Scientists, however, are obliged to
analyze problems and their causes from a wider point of view. In historical
geology, widening the perspective means expanding the window of time
by several million years (see Figure 3.1).
We have been in a warm epoch for some ten thousand years; this
has made possible the current expansion of humanity through the devel-
opment of agriculture. We should not forget, however, that this is
an interval within the cold period of the last million years. Moreover,
even within the last ten thousand years the climate has not been ab-
solutely uniform. There have been times when it was much colder
than now, and times as hot as or hotter than the present day, but these
smaller warm and cold cycles lasted only a few centuries, and were minor
in effect.
About 150 years ago we came out of a cold period known as the
Little Ice Age, which began in the 15th century and was a major influence
on many historical events. During the warm Middle Ages, on the other
hand, not only were wine grapes and other sensitive crops often grown
in regions of Great Britain that are now unsuited to them, but the

Arsuaga / Chosen Species 1405115327-4-003 Final Proof page 37 8.6.2005 5:23pm

Climate and Evolution

Average global











Absolute age
(millions of
of years)
4000 3000 2000 1000 570 225 65 1,7

Figure 3.1 Historical variation in global average temperature on earth. There

has been a steady decrease in average temperature since the Miocene, with marked
fluctuations in the Pleistocene which correspond to the ice ages

Vikings were able to colonize the southern coasts of Greenland for

several centuries, growing cereal crops, maintaining farms, and even
establishing a permanent episcopal diocese. The name itself, Greenland,
gives an indication of what the region was like. However, all traces of
the Vikings disappeared from Greenland at the beginning of the
16th century, when the wave of cold became so intense that in London,
King Henry VIII was able to cross the frozen Thames River in his
These small-scale fluctuations are nothing, however, compared with the
great climate changes which frame human evolution over the last four or
five million years. Over this period, continuing right up until today, we see
a general tendency toward a cooling of the planet, combined with an
overall decrease in precipitation. These phenomena have not occurred
uniformly, but have fluctuated in climatic cycles, progressively more
marked toward the present day.
Like the seasonal changes which take place each year, climate changes
affect regions in the high latitudes more than equatorial regions, where
much of human evolution occurred. In North America and Eurasia the
Ice Ages were very pronounced during the last million years, and during
this period great sheets of ice extended over a large part of these contin-
ents at regular intervals of about 100,000 years.

Arsuaga / Chosen Species 1405115327-4-003 Final Proof page 38 8.6.2005 5:23pm

Climate and Evolution

Paleotemperature Scales

The spectacular traces left by the ice during its advances and retreats in the
northern hemisphere have been used to establish a scale of changes in
temperature in Europe over time. Nevertheless the Ice Ages, although
they were large-scale climatic episodes, were also local phenomena which
affected different regions in different ways, even within the small Euro-
pean continent. Moreover, it is difficult to track them further back than
one million years.
The scales now used are based on marine paleotemperature curves,
derived from investigation of the sedimentary deposits accumulated in
the seabed. These marine records go back much further and are more
continuous than the continental records, and are a more reliable reflection
of global changes of temperature.
The most commonly used scale is based on the oxygen found in the
minute fossil shells of microorganisms known as foraminifera. These
minuscule protozoa live in the sea, and when they die their shells gradually
accumulate at the bottom. Dozens of test wells have been drilled in deep
water all over the world in order to obtain a microfossil sequence covering
a long period of time. In nature, oxygen occurs in two different isotopes,1
both of which are stable: oxygen-16, light and very abundant, and oxy-
gen-18, heavier and very rare. The ratio of the two oxygen isotopes in
seawater, and in the carbon dioxide dissolved in it, depends on the
temperature. Because this oxygen passes into the shells of the foraminifera
while they are alive, the temperature is recorded in their bodies for all
time; when they die this information is gradually deposited at the bottom
of the sea.
In order to relate these marine scales, which tell us the temperature of
the seawater, with continental climatic cycles, the deposits of dust trans-
ported from the land by the wind have also been traced in the seabed, in
order to see how these deposits have changed; it is assumed that the drier

Isotopes are atoms of a specific element which are chemically distinct from one
another. Because the small difference occurs in the nucleus of the atoms (in the number
of neutrons, to be precise), it has little effect on the normal properties of the element,
except for a minute difference in weight. The fundamental importance of isotopes, and
the reason they are so well known, is their nuclear properties: some are stable, like
carbon-12, and others are radioactive (i.e., they decompose naturally by emitting
radiation), like carbon-14. Some, such as uranium-235, are used to make atomic
bombs, while others, like uranium-238 and others, are not.

Arsuaga / Chosen Species 1405115327-4-003 Final Proof page 39 8.6.2005 5:23pm

Climate and Evolution

and more arid the adjacent continental regions, and the less vegetation
they have, the more dust will have blown toward the ocean. Fossil pollen
grains in the seabed have also been studied, in order to find out what kind
of plants covered the surface of the earth.
Marine records tell us what happened over the last few million years,
but they do not tell us why. What are the main factors contributing to the
changes in the earth’s climate?

Factors in Climate Change

The basic factors leading to climate change can be grouped into five main
categories: (a) catastrophic events; (b) geodynamic evolution of the
planet; (c) behavior of the hydrosphere-atmosphere system; (d) natural
fluctuations in the earth’s orbit around the sun; (e) the effect of the
biosphere, including human activity.
These factors, some of them very intricately enmeshed, produce very
different effects. Catastrophic events, which are sudden and unpredict-
able, such as the impact of giant meteorites or huge volcanic eruptions,
lead to marked changes of short duration; only if the change produced is
extremely drastic will it affect entire species.
At 10:00 a.m. on August 27, 1883, the largest recorded explosion in
history, much larger than any nuclear test, occurred in the volcanic crater
of Krakatoa (Indonesia). It was heard 3,500 km away; it destroyed an
entire island and spewed forth 21 sq m of rock. It shot enormous quan-
tities of gas and ash into the atmosphere, forming a vertical stream 80 km
high. The fine dust particles expelled into the stratosphere traveled
around the world several times and produced spectacular sunsets for
years, even in Europe. The resulting tsunami (‘‘tidal waves’’) traveled as
far as Hawaii and South America. More than 50,000 people died. It is
possible that the global temperature of the earth fell by half a degree, but
no long-term change occurred; no species disappeared. Currently there is
only one catastrophe theory which is taken seriously. This is the hypothesis
of a major meteorite impact, which is suggested as the cause of the
extinction of the dinosaurs – although this is still the subject of fierce
Geodynamic evolution includes a wide range of phenomena such as a
reduction in the flow of heat from the interior of the earth to the surface,
shifts in the geographic and magnetic poles, volcanic activity, and vertical

Arsuaga / Chosen Species 1405115327-4-003 Final Proof page 40 8.6.2005 5:23pm

Climate and Evolution

and horizontal movements of the earth’s crust. This last phenomenon is

of fundamental importance.
Over the last few million years continental drift has resulted in an entire
continent, Antarctica, being located exactly over the earth’s South Pole.
The snow that falls here thus accumulates, forming a layer of ice up to
4 km thick. In the northern, or Boreal, hemisphere the huge continental
masses of North America and Eurasia have also moved closer to the North
Pole. We are currently in a relatively warm period between two ice ages,
but during the ice ages the land at high latitudes was permanently covered
by ice, as around eighty percent of Greenland still is; the ice there reaches
depths of almost 3 km. At the North Pole there is no land on which snow
can accumulate, but as the Arctic Ocean is very enclosed, a permanent,
though not very thick, ice cap has formed which floats on the surface of
the ocean.
The hydrosphere-atmosphere system is extremely complex. Water’s
enormous capacity for storing heat means that the oceans act as vast
thermostats, moderating terrestrial fluctuations in temperature. In add-
ition, the seas largely control the amount of water vapor and carbon
dioxide present in the atmosphere. Precipitation (rainfall and snowfall)
depends on the amount of water vapor in the air; in addition, these two
gases are the main ones responsible for the so-called ‘‘greenhouse effect.’’

Milankovitch Cycles

The earth’s orbit around the sun results in two extremely regular tem-
perature cycles which we know well. The first is the alternation of day
(warm period) with night (cool period), caused by the rotation of the
earth about an imaginary axis passing through the North and South Poles;
as we know, this occurs every 24 hours. The second is the annual succes-
sion of seasons, which in the northern hemisphere gives rise to the series
spring (temperate), summer (warm), fall (temperate), and winter (cold).
This second cycle is due to the inclination of the earth’s axis of rotation
with respect to the plane of its orbit, an inclination which is currently
about 23.58.
There are four events which mark the beginning and end of the seasons,
as we know them in the temperate zones of the earth: two equinoxes
(spring, or vernal, and autumnal), and two solstices (summer and winter).
At the equinoxes day and night are of exactly equal length. In the
northern hemisphere, the summer solstice sees the day with the most

Arsuaga / Chosen Species 1405115327-4-003 Final Proof page 41 8.6.2005 5:23pm

Climate and Evolution

hours of daylight in the year, while the winter solstice sees the longest
night (see Figure 3.2). On the human scale, these are the only cycles we
perceive, and they are too short to leave geological traces.
The hypothesis that large-scale climate changes were due to natural
fluctuations in the earth’s orbit was first put forward in the 19th century.
However, the development of a quantitative theory relating the orbital
movements of the earth, the level of solar radiation, and the earth’s
climate was the impressive work of engineer Milutin Milankovic
(1879–1958), who dedicated over thirty years to the study of these
If the earth’s orbit around the sun was exactly circular (which it ‘‘vir-
tually’’ is), if the sun was exactly in the geometric center of the orbit
(which it ‘‘virtually’’ is), and the inclination of the earth’s axis of rotation

Daily rotation

Solar N N
radiation N


E quato r Equ a t o r
Day Night


Fall equinox

Winter solstice SUN Summer solstice

Vernal (spring)
equinox Trajectory of the
earth in its yearly orbit

Figure 3.2 The cause of the seasons. The inclination of the earth’s north–
south axis of rotation in relation to the plane of its orbit around the sun gives rise
to the seasons. In the northern hemisphere, the days are long and solar radiation is
strong in summer. In winter the days are short, and in addition solar radiation
meets the earth at a tangent (note that the opposite occurs in the southern
hemisphere). Note that the seasons are not directly related to the earth’s greater
or lesser distance from the sun

Arsuaga / Chosen Species 1405115327-4-003 Final Proof page 42 8.6.2005 5:23pm

Climate and Evolution

was always 23.58 (as is ‘‘virtually’’ the case), there would be no large-scale
climate changes caused by fluctuations in the level of solar radiation: all
summers would have been the same for thousands of millions of years.
However, all those ‘‘virtually’’s added together, along with some others
we have not mentioned, result in very gradual changes in the amount of
solar radiation reaching the earth each year. According to Milankovic’s
theory, this is the slow but inexorable motor driving climate change.
In addition to equinoctial precession (Figure 3.3), which results in
periods of very hot summers alternating with periods of temperate

Present Vernal (spring) equinox, March 20




152 million km 147 million km Perihelion
Empty focus
Center of Focus
ellipse Winter solstice, December 21

In 11,000 years


LL Fall equinox
Fall equinox, FAL
September 22



Aphelion Perihelion




Vernal equinox SPR

Figure 3.3 Precession of the equinoxes. The position of the equinoxes and the
solstices which mark the beginning of the seasons changes slightly each year.
Currently, winter in the north begins when the sun is closest to the earth (the
perihelion); winters are therefore not particularly cold. In about 11,000 years’
time the winter will occur when the sun is farthest away (the aphelion), and
northern winters will be harsher. Thus equinoctial precession periodically tempers
or heightens the intensity of the seasons

Arsuaga / Chosen Species 1405115327-4-003 Final Proof page 43 8.6.2005 5:23pm

Climate and Evolution

summers approximately every 11,000 years, Milankovic looked at two

other, slower fluctuations in his paleoclimatic analysis. The first of these
was the change in the inclination of the earth’s axis of rotation relative to
the plane of its orbit, which is currently around 23.58, and varies between
218 and 24.58 over a period of about 41,000 years. The second, the
slowest of all, with cycles of between 100,000 and 400,000 years, is
related to changes in the shape of the orbit (whether it is more rounded
or more elongated). All of these orbital variations combine to give rise to
complex fluctuations over time in the amount of solar radiation the earth
Since the development of marine paleotemperature scales in the 1970s,
particularly the scale based on oxygen isotopes, we have undeniable
evidence that climate fluctuates in cycles which coincide with those
predicted by Milankovic’s theory. All we are now missing is detailed
knowledge of the mechanisms which amplify the slight variations in
solar radiation levels predicted by the theory, resulting in major changes
in climate.

A Climatic Model for Equatorial Africa

At the present time rainfall in subtropical Africa is markedly seasonal,

and follows the yearly cycles of the African monsoon. During the northern
summer the land in the interior of the continent is heated up, attracting
moisture-laden air from the equatorial Atlantic. This results in abundant
rainfall in the western and central regions of subtropical Africa. In
eastern Africa rainfall is always much lower, because in addition to the
reduction we would expect owing to the greater distance which the
clouds have to travel, there is a mountain barrier which prevents
them from reaching this region. In winter atmospheric circulation is
reversed, and cold, dry winds from the northeast blow over the whole
of the region. Once again, the effects are more serious in eastern Africa; in
the west, local currents continue to bring warm, moist air from the
As a result of these conditions, the vegetation in western subtropical
Africa still consists of humid forest today. East Africa, on the other hand, is
much more arid. It now has savanna-type ecosystems, with grassland
predominating over trees. We shall return to this geographic factor, the
cause of such a marked division between the eastern and western regions

Arsuaga / Chosen Species 1405115327-4-003 Final Proof page 44 8.6.2005 5:23pm

Climate and Evolution

of subtropical Africa, when we look at what is known as the ‘‘East Side

Story’’ hypothesis.
Marine records show that the seasonal summer monsoon already
existed 5 million years ago; however, more favorable overall conditions
in terms of temperature, humidity, and atmospheric levels of carbon
dioxide (CO2 ) meant that more or less humid forest covered the whole
of subtropical Africa.
Peter deMenocal has constructed a theoretical model of how different
factors have affected climate in the low latitudes of Africa over the last
few million years. The model explains how the climate of subtropical
Africa can be influenced by climate fluctuations in the north – ice ages.
Among other elements, it studies the effect of cooling of the north
Atlantic on the monsoon in Africa. According to deMenocal, around
2.8 million years ago a phenomenon occurred which changed the climatic
history of the northern hemisphere decisively and had a major impact
on the ecosystems within which our ancestors were evolving: this was
the beginning of large-scale climate fluctuations, with permanent ice
over much of the north during the cold epochs. Marine records in
subtropical zones of the Atlantic and Indian Oceans, which surround
the African continent, show that these climate fluctuations had a marked
influence on the climate in equatorial Africa. Since this time the retreat of
the forest in this region, giving way to savanna and grassland, has been
unstoppable. It seems that this was the situation between 2.8 and 1 million
years ago.
Marine records also show that the influence of fluctuations in the
north on the climate of this region of Africa has become stronger over
the last million years. DeMenocal’s theoretical model can explain the
increase in aridity in subtropical Africa and its relation to northern ice
ages. The main reason for this, it is suggested, is that the cooling of
the waters of the north Atlantic led to a series of cooler, drier African
summer monsoons which destroyed the delicate tropical forests of
East Africa.
Nevertheless, since astronomical factors have always existed, and ice
ages have only occurred in particular eras of the earth’s history, we have
to recognize that such extreme situations only occur when other circum-
stances are superimposed on orbital fluctuations. One of these, of course,
is the poleward movement of the continental landmasses, which encour-
ages the accumulation of ice and hampers the movement of warm tropical
waters. The other is changes in the atmosphere, and most particularly, the
influence of CO2 .

Arsuaga / Chosen Species 1405115327-4-003 Final Proof page 45 8.6.2005 5:23pm

Climate and Evolution

The Controversial Gas

The natural presence of water vapor (H2 O), carbon dioxide, and other
atmospheric gases ensures that the average temperature of the earth’s
surface is 158C, rather than 158C: if it were not for the natural green-
house effect, the earth’s surface would be a layer of ice! For this reason,
when we talk of the greenhouse effect as something potentially danger-
ous, we should really use the term ‘‘heightened greenhouse effect’’ – in
other words, the additional overheating which can be caused by emissions
resulting from human activity.
CO2 is the slightly sour, usually harmless gas that gives us the bubbles in
our sodas. It makes up only a very small proportion of the atmosphere,
barely 0.03 percent, but it is nevertheless vitally important: not only is it
the main agent in the beneficial greenhouse effect (when this is not
magnified by human activity), it is also the basic source of organic
carbon – the carbon of which all living beings are made.
With the help of sunlight, plants convert water and CO2 into organic
matter through photosynthesis. Most plants belong to a group known as
the C3 plants, because they fix carbon dioxide through a mechanism
which uses a molecule with three carbon atoms. A minority of plants,
mostly grasses with tough, fibrous stems, belong to another group known
as the C4 plants, since their mechanism for fixing CO2 uses a molecule
with four carbon atoms. A few cultivated crops, such as maize and sugar
cane, belong to this second group.
With the right humidity and temperature conditions, and with abun-
dant CO2 , C3 plants develop much better than C4, but with the current
level of CO2 in the atmosphere, which is very low compared with other
geological epochs, C3 plants find it difficult to survive in hot, dry envir-
onments. Today, C3 plants still predominate in the temperate and cold
climates. In the hot climates of equatorial latitudes, C3 plants grow in
abundance in the humid forests and also in the so-called rain forests
(where an unlimited amount of water is available). The C4 plants, on
the other hand, are grasses and reeds typical of the open, dry, sunny
environments grazed in Africa by zebra and antelopes such as gazelles,
hartebeest, impala, and gnu, and by elephants, hippopotamus, and other
herbivores of the great savannas. As these C4 plants have fibrous, mineral-
rich stems they cause much wear on the teeth of the mammals which eat
them; for this reason their teeth have high crowns, so that they will last a
long time.

Arsuaga / Chosen Species 1405115327-4-003 Final Proof page 46 8.6.2005 5:23pm

Climate and Evolution

In addition to their appearance, C3 plants are chemically distinct from

C4 plants, since the C4 plants contain a larger proportion of a particular
rare isotope of carbon (carbon-13). Thure Cerling and his colleagues have
carried out an exhaustive analysis of the quantity of this stable but rare
isotope of carbon in the enamel of fossil teeth from grass-eating animals,
particularly Equidae (the group which now consists of horses, zebras, and
their relatives, the donkeys and wild asses), but also Proboscidea (the
elephant group), some extinct South American mammals known as
notoungulates, and other groups of large herbivores. The regions studied
were Europe, East Africa, Pakistan, North America, and South America,
and the period covered was the last 20 million years. The researchers
found that 8 or more million years ago levels of this rare isotope were
low in all groups, indicating a world dominated by C3 plants. However,
2 million years later the situation began to change in East Africa, Pakistan,
equatorial and South America (in Europe and Africa C3 plants have always
been dominant).
Cerling and his colleagues concluded that between 8 and 6 million years
ago the concentration of CO2 in the atmosphere began to decrease – a
decrease which has continued up to the industrial era. This resulted in the
expansion of open ecosystems dominated by C4 plants, and the reduction
of the forested areas. Cerling and his colleagues make the interesting
observation that the changes in plant cover which affected extensive
regions of the earth were accompanied by major changes in fauna, with
the expansion of mammals adapted to open environments. These herbi-
vores, with their high dental crowns suitable for grazing grass, replaced
those which browsed the trees.
One final note. The reader will no doubt have realized that the emission
of CO2 into the atmosphere which humans cause by burning fossil fuels
(coal, oil, etc.) will also have ecological effects in the future, favoring C3
plants. This is just one of the factors, combined with the heightened
greenhouse effect, the hole in the ozone layer, acid rain, and others, on
which we are exerting an irresponsible influence, well before we have
managed to understand the complex mechanism of the earth’s climate.

The End of Paradise

In ecological terms, the progressive decrease in the volume of CO2 in the

atmosphere, combined with climatic factors, led to the belt of hot tropical

Arsuaga / Chosen Species 1405115327-4-003 Final Proof page 47 8.6.2005 5:23pm

Climate and Evolution

Figure 3.4 The giant gelada Theropithecus oswaldi (right), shown on the
same scale as a female hominid, Paranthropus boisei

forest which extended over much of the Old World becoming fragmented
and declining from the end of the Miocene, and particularly during the
Pliocene and Pleistocene. This loss of habitat probably resulted in the
disappearance of many species of hominoids, although this may not have
been the only cause of the reduction to the very narrow range which exists
today. In the Miocene there were various species of hominoids living
within the same region, whereas today no more than two species live
together. Another major cause of the decline may have been ecological
competition with the other Old World monkeys, the Cercopithecidae,
which are much more abundant and varied today.
However, this same climate change led to the appearance and spread
of more open ecosystems over much of Africa during the late Miocene
and Pliocene, with new species of plants and animals. Among these the
hominids (our ancestors and our closest relatives) soon appeared, as
we shall see in the next chapter. The ancestors of the patas monkey, and
of the baboons and the geladas, also took advantage of this change in
environment. One form of gelada, Theropithecus oswaldi, developed to
enormous size during the Pleistocene (weighing up to 100 kg in excep-
tional cases), and lived in the same zones as humans, who may even have
hunted them and contributed to their becoming extinct (Figure 3.4).
Numerous remains of this giant monkey, together with an enormous

Arsuaga / Chosen Species 1405115327-4-003 Final Proof page 48 8.6.2005 5:23pm

Climate and Evolution

quantity of stone tools, have been found in the Olorgesailie deposit in

Kenya (about 800,000 years old).
As we see, we are far from the only primates to have come down from
the trees, or to put it another way, to adapt to a world in which trees had
been replaced by grass, and forests by grassland.

Arsuaga / Chosen Species 1405115327-4-004 Final Proof page 49 4.6.2005 2:36pm

The Origin of Humanity

So for every man who has ever lived, in this universe, there shines a star.
Arthur C. Clarke, 2001: A Space Odyssey

Molecular Clocks

Studies in molecular biology indicate that our line separated from the
chimpanzee line between 4.5 and 7 million years ago – in other words, at
approximately the same time that the gradual decrease in atmospheric
CO2 levels was beginning to affect the African ecosystems, as we saw in
Chapter 3. This coincidence makes it tempting to speculate that hominids
originated as a direct result of ecological change and the spread of open
environments, to which they adapted from the beginning. However, as we
shall see later, it now seems that the oldest representatives of our group,
the first hominids, were as much forest-dwellers as are today’s chimpan-
zees, and that the gradual adaptation to drier, less densely forested envir-
onments occurred later.
However this may be, molecular biologists have calculated this
interval between 4.5 and 7 million years using their molecular clocks.
The basis of these molecular clocks is that the genetic difference between
species, such as our own and the chimpanzees, ought to be related to the
time which has passed since the two lines separated. In other words,
genetic divergence increases with time, as does the morphological differ-
ence between two lineages which diverge from one another to follow
different evolutionary paths.

Arsuaga / Chosen Species 1405115327-4-004 Final Proof page 50 4.6.2005 2:36pm

The Origin of Humanity

But this assertion that genetic difference is related to the length of time
the lines have been separated is only valid if appropriate genes are chosen
for analysis. The genes suitable for use as molecular clocks are those
known as ‘‘neutral,’’ those which confer neither advantage nor disadvan-
tage, and on which natural selection therefore does not act. In neutral
genes, the naturally occurring spontaneous mutations accumulate at a
constant rate, without being either eliminated or favored, like snow falling
On the other hand, the ‘‘non-neutral’’ genes that selection does target
may be modified at different, varying rates depending on the intensity
of the pressure for selection exerted upon them. In other words, if a
particular gene (the correct technical term is an allele) is very beneficial
for the individual that carries it, it is sure to spread rapidly throughout
the population. If, however, it confers a disadvantage, its frequency of
occurrence in the population will fall rapidly because it has natural
selection, a powerful enemy, against it. Furthermore, what is beneficial
today may not be so tomorrow, or may not be so in another species – so
these non-neutral genes are of no use for measuring time in evolution.
To take an example from everyday life, the power and capacity of
personal computers, which are subjected to market pressure for selection,
increase very rapidly, and not at a constant rate. This is a clock which
gains time.
But in order to calculate the rhythm of change of neutral genes, what
is known as the rate of mutation, we have to return once again to
fossils, measuring the genetic difference between two species for which
we know, from fossils, how long their lines have been separated. For
example, in order to work out how long it is since the human and
chimpanzee lines separated, we can use the human/orangutan pair.
Measuring the genetic distance between the two is the easy part, although
since not every gene can be used the calculations vary depending on
which ones are chosen. Establishing when the orangutan line separated
is a very different matter. Sometimes the figure of 13 million years is
used: this corresponds to the first fossils attributed to the species
Sivapithecus, which themselves are believed to mark the beginning of the
evolution of orangutans.
In other words, in order for the ‘‘molecular clock’’ to work, many
things are needed: genes which natural selection does not ‘‘see’’ but
which are known to us, constant rates of mutation, and a good paleonto-
logical reference point – too many things – but nevertheless, this interval
of between 4.5 and 7 million years for the human/chimpanzee separation
is acceptable to paleontologists, as we shall see shortly.

Arsuaga / Chosen Species 1405115327-4-004 Final Proof page 51 4.6.2005 2:36pm

The Origin of Humanity

The First Fossil Hominids

Before continuing we should pause for a moment to clarify a question of

terminology, in order to avoid confusion. Some paleoanthropologists use
the term ‘‘hominid’’ in a very broad sense to refer to humans, chimpan-
zees, gorillas, and the fossil relatives of the entire group. We prefer to use
the term hominid in its more traditional sense, to refer only to modern
humans and the fossils of our own evolutionary line, i.e. the species which
emerged after the separation of the chimpanzee line. Other authors define
hominids as bipedal primates. However, although it is true that all the
species with erect posture come under our definition of hominid, as we
shall see, we do not yet know for certain whether the first hominids
already walked on two feet. All bipeds are hominids, but it may be that
not all hominids were bipedal.
A fragment of a mandible (lower jawbone) with a molar found at
Lothagam (Kenya), to which has been ascribed a geological age of more
than 5.6 million years, may be that of a hominid, although the small size
of the preserved fragment makes it difficult to be sure. There are other
fossils which similarly offer little information, including a jawbone frag-
ment from Tabarin and a fragment of the proximal (upper) humerus from
Chemeron (both in Kenya), dated at about 4.5 million years old.
Aside from these isolated, doubtful remains, the oldest set of hominid
fossils yet found was located in 1992 by a team led by Tim White, Gen
Suwa, and Berhane Asfaw in the middle reaches of the Awash River in the
Afar region of Ethiopia. These Middle Awash fossils have been published
only in part, although White and his colleagues have already created a new
genus and species for them: Ardipithecus ramidus (the words ardi and
ramid come from the Afar language and mean, respectively, ‘‘ground’’
and ‘‘root,’’ while pithekos means ‘‘ape’’ in Greek). The information that
is available on these fossils indicates that they are very primitive forms of
hominid, around 4.4 million years old. In fact, they show features so
primitive, particularly in their dentition, that it seems likely that they
cannot be very far from the division between the chimpanzee and
human lines. An age of between 4.5 and 7 million years therefore seems
acceptable for the time being, and if we were pinned down we would go
for a date closer to 4.5 than to 7 million years. In any case, it is likely that
we will have a definitive answer before very long.
It also appears that Ardipithecus ramidus lived in a forest environment.
This conclusion is drawn first from the type of mammals – forest-dwellers –
with which the hominid fossils appear. Monkeys of the colobus type and

Bahr el Ghazal Australopithecus bahrelghazali

Hadar Australopithecus afarensis

Middle Awash Ardipithecus ramidus

Australopithecus afarensis

Lake Turkana Australopithecus anamensis

Laetoli Australopithecus afarensis

Arsuaga / Chosen Species 1405115327-4-004 Final Proof
page 52

Makapansgat Australopithecus africanus

Sterkfontein Australopithecus africanus

4.6.2005 2:36pm

Taung Australopithecus africanus

Figure 4.1 Location of the main deposits containing fossils of Australopithecus and Ardipithecus
Arsuaga / Chosen Species 1405115327-4-004 Final Proof page 53 4.6.2005 2:36pm

The Origin of Humanity

tragelaphine antelopes (the group of the kudus, sitatungas, and others

with spiral horns) are especially abundant. Secondly, the outer enamel
layer on the teeth of Ardipithecus ramidus is thin, as in chimpanzees,
which eat fruit, leaves, tender stalks, shoots, and other soft plant products.
However, the teeth of fossil hominids dating from after Ardipithecus
ramidus have a thick layer of enamel, protecting them from the wear
caused by a diet incorporating tough plant elements such as roots, tubers,
grains, nuts, and so on. So it appears that the first ancestors of man, the
first hominids, were primates who lived in the forest and had a diet very
similar to that of modern chimpanzees. But we shall leave the detailed
analysis of hominid diet for a later chapter.
On the basis of some aspects of the base of the skull, in fairly fragmen-
ted remains, White and his colleagues have suggested that these first
hominids were bipedal and walked as we do. However, this remains to
be demonstrated on the basis of hip and leg bones, some of which are
known to have been found in the most recent excavations.

Change of Habitat

A Kenyan team led by Meave Leakey (wife of the famous fossil-hunter

Richard Leakey, of whom more later), found hominid fossils about
4 million years old (between 3.9 and 4.2 million years old, to be precise)
in Kanapoi and Allia Bay, either side of Lake Turkana in Kenya. In 1995, on
the basis of these, they identified a species they named Australopithecus
anamensis (the term anam means ‘‘lake’’ in the Turkana language, so the
name of this species could be translated ‘‘australopithecine of the lake’’).
These are again very primitive hominids, to judge from a maxilla (upper
jawbone) and a mandible (lower jawbone) which have been found,
but they have thick enamel on the molars. The association of fossils in
which they occur suggests an open forest environment, or more or less
wooded savanna with streams. There are also colobus monkeys and spiral-
horned antelopes, but these are associated with other species more com-
mon to open environments, such as jerboas, a type of mouse from the
arid steppes.
The hominid remains found include a tibia (shinbone), which its dis-
coverers believe indicates that these primates were bipedal. We are wit-
nessing for the first time, about 4 million years ago, the appearance of
hominids who have begun to change their way of life, their environment,
and their diet significantly, and who also move in a completely new way.

Arsuaga / Chosen Species 1405115327-4-004 Final Proof page 54 4.6.2005 2:36pm

The Origin of Humanity

Although hominids are not the only primates living in the savannas and
grasslands, the erect posture is a completely new innovation.
The next million years (roughly speaking) correspond to another East
African species, known as Australopithecus afarensis (which can be trans-
lated as ‘‘australopithecine of the Afar region’’). Most of the fossils of this
species have been found in the Hadar area, on the lower reaches of the
Awash River (in the Afar region, Ethiopia), and in Laetoli (Tanzania).
The Laetoli fossils, which include a mandible classified as the type
specimen (holotype) of the species (known as L.H. 4), are dated at 3.5
million years old; the Hadar fossils are between 3 and 3.4 million years
old. Donald Johanson’s team, working in Hadar since 1972, has discov-
ered numerous remains, so that Australopithecus afarensis has a reasonably
complete fossil record. This includes the skull of a male (A.L. 444-2),
discovered by Yoel Rak in 1992 (Figure 4.2), and a large part of the
skeleton of a female (A.L. 288-1), known worldwide as Lucy, as she was
named by Johanson when he found her in 1974.
An assemblage of fossils which includes the most complete known
mandible of this species was found in the Maka deposit, 3.4 million
years old, on the middle reaches of the Awash River. Some teeth a little
over 4 million years old found in Fejej, southern Ethiopia, and a fragment
of frontal bone from Belohdelie in the Middle Awash, approximately 3.9
million years old, have also been attributed to Australopithecus afarensis.
However, the assignation of these fossils to the species is subject to review
and they could, given their chronology, belong to Australopithecus
The associations of vertebrates in the Hadar deposits suggest changes in
the paleoecological conditions over the 400,000 years of geological his-
tory recorded there. Australopithecus afarensis appears to have lived both
in a fairly dry forest environment and in a cool savanna with corridors of
forest along the river valleys – in other words, neither a humid forest nor
an arid steppe, but rather an intermediate habitat.
One of the fundamental problems of paleontology is that of grouping
fossils into species, since unfortunately the remains do not appear with
labels in the deposits. It is the paleontologist’s job to find their place in
Living species which are related to one another are often morphologic-
ally very similar, or may differ only in external features such as color, fur,
behavior, and other characteristics which, although they may be very
striking, are not reflected in the skeleton, the part that becomes fossilized.
Ian Tattersall has pointed out that many modern species of primates
would not be recognized as different species if we only considered their

Arsuaga / Chosen Species 1405115327-4-004 Final Proof page 55 4.6.2005 2:36pm

The Origin of Humanity

Sagittal ridge

Nuchal ridge

Figure 4.2 Male Australopithecus afarensis

Arsuaga / Chosen Species 1405115327-4-004 Final Proof page 56 4.6.2005 2:36pm

The Origin of Humanity

skeleton; we may thus be seriously underestimating the number of fossil

species, since we might group together, as one species, two species which
had very different external features in life, although they had an identical
or very similar skeleton.
Conversely, great variation can occur within one species when there are
major differences between males and females. This differentiation be-
tween the sexes is known as ‘‘sexual dimorphism,’’ and can affect size,
shape, or both. The paleontologist could thus make the mistake of assign-
ing to two different species fossils which simply represent different sexes
of the same species.
In short, these problems are often the subject of major debates among
specialists. Australopithecus afarensis is no exception, and when the species
was identified in 1978 by Donald Johanson, Tim White, and Yves Cop-
pens, some researchers did not accept that all the fossils from Hadar and
Laetoli should be grouped together in a single species – albeit a very
variable one with major sexual dimorphism. Many paleoanthropologists
saw two species in this set of fossils, rather than just one, although they
did not agree on which fossils should be allocated to which species. Earlier
we mentioned the skull A.L. 444-2, which is large and could be a male of
the same species as Lucy, a small individual who may have been female.
The alternative is that these represent two different species. We chose this
example deliberately because, as often happens in paleontology, two dif-
ferent parts of the skeleton are being compared: the skeleton of the body
(or postcranial skeleton) of A.L. 444-2 is not preserved, while only a small
part of Lucy’s skull has been recovered. In addition, Lucy is 3.2 million
years old, while A.L. 444-2 is around 200,000 years younger.
Some experts have found significant differences in the postcranial skel-
eton, and on this basis they identify a completely bipedal species at Hadar
and Laetoli, one they suggest is directly related to us and our species,
rather than being an ancestor which combined the capacity to walk on
two legs with the ability to climb trees. Finally, some people have even
suggested that the females were lighter and were climbers, while the males
were heavier and bipedal!
There is one very convincing argument for the unity of Australopithecus
afarensis, based on the fossils found at Hadar site A.L. 333. Numerous
hominid remains from 3.2 million years ago have been found here, with
virtually no bones of other animals. These hominid fossils represent at
least thirteen individuals of different ages, who might have died together
in a natural disaster such as a flood. It is very possible that they formed
part of the same group and therefore the same hominid species. In fact,
this handful of fossils is known colloquially as the ‘‘first family.’’

Arsuaga / Chosen Species 1405115327-4-004 Final Proof page 57 4.6.2005 2:36pm

The Origin of Humanity

If the sample from deposit A.L. 333 had included only large individuals,
or only small ones, or individuals with a particular morphology, then
we might conclude that fossils of different species had been artificially
included in Australopithecus afarensis. Conversely, if the sample included
all the sizes and morphotypes found at Hadar and Laetoli, we could be
sure that Australopithecus afarensis is a real species and not a hodgepodge
of odds and ends. In fact, there is wide variation within the A.L. 333
sample – as wide as has been suggested by the researchers who identified
Australopithecus afarensis.

East Side Story

We have discussed what is known about when hominids first appeared;

now we need to examine where this happened: where was our first home?
As we have seen, the fossils of the first hominids were found in East Africa.
More specifically, these fossils were found along the Great Rift Valley,
an enormous, widening fracture in the earth’s crust which runs from
Mozambique, through Malawi, the Great Lakes region, the Afar region
of Ethiopia, and the Red Sea to arrive at the Dead Sea, between Israel
and Jordan.
The geographical distribution of the first hominid fossils suggests that
our group originated in East Africa – this is the theory that Yves Coppens
has called ‘‘East Side Story.’’ According to this hypothesis, throughout
the Miocene there was a great belt of tropical forest extending from the
Gulf of Guinea across to the Indian Ocean. The tectonic process which
formed the great, continent-spanning fracture of the Great Rift Valley
must have brought with it changes in relief, raising great mountain
barriers and high plains. From the end of the Miocene these separated
the eastern ecosystems, with their increasingly open environments inhab-
ited by hominids, from the western ecosystems, which had humid forest
environments and were populated by the ancestors of chimpanzees
and gorillas.
This hypothesis becomes even more attractive when we consider that it
would mean that our emergence was not a unique phenomenon. We are
simply part of a community of animal and plant species – a biota, to use
the biogeographical term – characteristic of an entire region, and linked to
the geological and climatic history of that region. Now, although this
hypothesis appears very reasonable, it starts from the basis that the first
hominids originated in East Africa. It is in this region that the oldest fossils

Arsuaga / Chosen Species 1405115327-4-004 Final Proof page 58 4.6.2005 2:36pm

The Origin of Humanity

known up to this day have been found. However, at the end of 1995
Michel Brunet and his colleagues published the discovery north of N’Dja-
mena, in Chad, of the front portion of an australopithecine mandible and
an isolated premolar, which are dated to between 3 and 3.5 million years
old on the basis of the accompanying fauna. The authors suggested, in a
subsequent study, that this was a species different from Australopithecus
afarensis, which they named Australopithecus bahrelgazali (the term bah-
relgazali refers to the region of Chad in which the fossils were found –
Bahr el ghazal, which in Arabic means ‘‘river of the gazelles’’). This
discovery suggests that, early on in their history, hominids spread far to
the west of their East African birthplace – if this is in fact where they
The African deposits we have discussed thus far, which are lacustrine or
fluvial sediments (from lakes or rivers), usually contain aquatic species
such as turtles, crocodiles, fish, and hippopotamus. This does not help
us to determine in what kind of ecosystem the hominids lived: other
animals, not aquatic or amphibian, are more useful to us. This is no simple
matter, for a sedimentary basin may contain the remains of animals that
lived and died in very different environments, and have been transported
there by streams and rivers. Thus, everything accumulates at the bottom
of the basin, creating many problems for paleontologists, who try to
resolve them principally on the basis of common sense. The application
of common sense to the study of the formation of such deposits consti-
tutes a paleontological discipline in itself, a very important one, known as
taphonomy. Using taphonomy we can establish, for example, whether a
bone has been transported over a long distance or whether the animal
died close to where it is found. Fortunately, the study of the adaptations
found in fossil species, or functional paleomorphology, also contributes to
determining the place of these species in the ecosystem (their niche), and
what their environment was like.
We shall return to the question of the environment in a later chapter
when we look at the diet of the first hominids. For now let us take a
moment to consider the fossils associated with Australopithecus bahrelga-
zali. Michel Brunet imagines that these australopithecines lived in a
variety of environments, including forest corridors (characterized by the
presence of bushpigs), wooded savanna with elephants, and grasslands
where rhinoceros grazed.
The main differences between the Chad fossil and Australopithecus
afarensis are in the interior face of the symphysis, or anterior (front)
part of the jawbone, which has a fairly flat and vertical surface, without
the strong transverse reinforcements or tori (ridges – from the Latin torus)

Arsuaga / Chosen Species 1405115327-4-004 Final Proof page 59 4.6.2005 2:36pm

The Origin of Humanity

characteristic of other australopithecines. In this respect Australopithecus

bahrelgazali is similar to our genus, the genus Homo. However, all the
premolars have three roots – a primitive trait. To further complicate
matters, the jawbone of Australopithecus bahrelgazali has wide premolars.
This expansion of the premolars is, as we shall see below, typical of some
later hominids known as Paranthropus.
This combination of features may not constitute sufficient grounds for
considering the Chad fossil as a different species, but if it does, we are
presented with a new situation. For the first time in our evolutionary
history, two species of hominids would coexist, albeit in different regions.
While human evolution has traditionally been explained as a linear suc-
cession of species, we shall see that the evolutionary tree in general – and
our case is no exception – has many branches, even when in some cases
only one branch (like ours) ultimately reaches the present day.

Dating Fossils

The reader has perhaps already wondered how we know the age of these
fossils of our very distant ancestors. The paleontologist Yves Coppens,
who gives many lectures, is often asked this question, to which he recom-
mends the reply: ‘‘Trust us, we know the age of fossils, we have methods
for determining it, we are professionals.’’ At the risk of boring the reader,
we shall disregard his advice and attempt a brief discussion of this funda-
mental question.
Because of the intensity of the internal forces which come into play,
fracturing of the earth’s crust is often accompanied by volcanic activity,
which may cause ash to be thrown up into the air during the course of
eruptions. Winds and water transport the ash, which is finally deposited in
beds intercalated between the layers of sediment which contain the fossils.
These layers of volcanic ash, or tuffs, are very useful in correlating and
dating formations. Even two successive eruptions from the same volcano,
with very little time between them, have distinct characteristics, known as
their ‘‘chemical fingerprints.’’ Using chemical analysis, we can compare
two layers of volcanic tuff and find out whether they are the same, even if
they do not run continuously in the field because the rocks are fractured
into blocks, as is typical of the geology of rift valleys.
During the 1980s a method of dating these tuffs was invented, based
on melting a single small crystal of a mineral of the potassium feldspar
group using a laser beam. When the crystal melts, the laser releases a given

Arsuaga / Chosen Species 1405115327-4-004 Final Proof page 60 4.6.2005 2:36pm

The Origin of Humanity

quantity of the gas argon, which is measured with an instrument known as

a mass spectrometer. The isotope argon-40 derives from the decay of a
radioactive isotope of potassium (potassium-40) contained in the mineral.
When the mineral was formed it contained only potassium and no argon.
Since radioactive decay occurs at a known, constant rate, the final ratio of
radioactive potassium to argon gives us a very reliable age for the tuff.
A variant of this technique, known as argon-39/argon-40 dating, is what
is now used.
Another method used to date volcanic rocks is the fission-track method.
The decay (fission) of radioactive uranium (uranium-238) produces marks
(tracks) in the crystals of certain minerals, such as zircon. The density of
these tracks depends on the quantity of uranium in the mineral and the
time elapsed since the volcanic eruption in which the mineral formed.
The fossils lying between successive tuffs in sedimentary sequences can
be dated with a precision unimaginable a few years ago. The Middle
Awash is an extremely fortunate case, where the majority of the fossils of
Ardipithecus ramidus come from sediments enclosed, like a geological
sandwich, between two volcanic tuffs, one below and one above. Both
tuffs are approximately the same age, 4.4 million years old, and this is also
the age of the fossils found between them (unfortunately, this is not the
general rule, and often there are hundreds of millions of years between
tuffs lying above and below fossils).
The argon-39/argon-40 and fission-track methods are used with vol-
canic materials which, although frequent in East Africa, are not found in
by any means all of the formations containing hominids. Other radiomet-
ric methods use other isotopes, like carbon-14 or the uranium series. The
carbon-14 method (the first to be developed) can be used only for organic
matter (matter of animal or plant origin), and is very reliable. Unfortu-
nately, even with the latest improvements, it cannot date anything older
than 50,000 years.
Caves frequently contain speleothems (stalactites and stalagmites),
which form through continuous precipitation of calcium carbonate dis-
solved in water. If we are lucky and the carbonate crystals are sufficiently
pure, speleothems can be dated using the uranium series method, up to a
maximum of about 350,000 years.
But in many cases we do not even have datable speleothems in the
formations. Two related techniques, known as electron spin resonance
(ESR) and thermoluminescence (TL), have been developed for these
cases. ESR is usually used on the tooth enamel of various mammals,
while TL dating is applied to charred flint tools, and a type of sediment
which has been exposed to sunlight. The basis of both techniques is that

Arsuaga / Chosen Species 1405115327-4-004 Final Proof page 61 4.6.2005 2:36pm

The Origin of Humanity

minerals such as flint or quartz, like teeth and bones, act as natural Geiger
counters, accumulating the radiation received over time.
Another method used in the measurement of geological time (geo-
chronology) is paleomagnetism. The earth acts as a magnet, creating a
geomagnetic field around it, with two poles (north and south). This field
orients the compass needle, indicating the position of the magnetic
north pole, which is today close to the geographic North Pole. But it
also orients ferrous clay minerals, like tiny compasses, provided that
they are deposited slowly in a calm environment, such as an undisturbed
pool or lake.
Over extremely long periods, the magnetic poles exchange position (so
that the magnetic north pole is located close to the geographic South
Pole). The location of the poles at any given moment is recorded in the
clay minerals making up the layers of sediment. These changes in the
earth’s magnetic polarity have been dated, enabling us to establish a
scale showing the sequence of alternating periods of one or other type
of polarity. Each of these long periods of time is called a chron. A subchron
is a shorter unit of time, with a polarity opposite to that of the chron
within which it occurs; even shorter periods are known as excursions.
Paleomagnetism cannot give us an absolute date for a formation, but it
can help with dating when combined with other methods.
In addition to these methods of dating, the fossils of the animals
associated with the hominids also help to establish their relative age,
since the evolution of species means that animals, including hominids,
change over time. Thus biochronological scales can be created, which can
be calibrated with the absolute dating obtained by physical methods.

The Taung Child

November 28, 1924 was a great day in the history of paleoanthropology.

On that day Raymond Dart (1893–1988), a young professor of anatomy
at the University of Witwatersrand in Johannesburg, South Africa, re-
ceived a package sent from the Taung quarry containing a child’s skull, in
which Dart recognized a very remote ancestor of ours. He created a new
species and a new genus for this creature: Australopithecus africanus (we
have waited until now to explain the meaning of the term australopithecus,
which is made up of the terms pithecus, or ‘‘ape,’’ and austral, meaning
‘‘south’’). The formation where the skull was found was destroyed
and did not yield any further hominid fossils, but two other quarries,

Arsuaga / Chosen Species 1405115327-4-004 Final Proof page 62 4.6.2005 2:36pm

The Origin of Humanity

Sterkfontein and Makapansgat, proved very ‘‘productive’’ of fossils of

Australopithecus africanus.
Thanks to the Sterkfontein Formation, we have an extensive record of
Australopithecus africanus (Figure 4.3), including a very complete and
emblematic skull (found in 1947), referenced as Sts 5, and popularly
known as Mrs Ples (Ples being an abbreviation of Plesianthropus, the
genus to which this specimen was at first ascribed, although it was later
realized that it was identical with Australopithecus). Another important
skull is Sts 71, also found in 1947; in 1989 a very complete skull (Stw
505), apparently male, was recovered, but this has not yet been published
in detail. Many pieces of postcranial skeleton have also been recovered, the
best known being skeletons Sts 14 (discovered in 1947) and the recently
discovered Stw 431. In fact these South African limestone quarries are
caves filled with fossils and a much-hardened sediment, forming a very
hard breccia (coarse-grained rock) which renders the extraction of fossils
extremely difficult.
In the region of South Africa where all of these fossils were found there
are no layers of volcanic ash to enable us to date the fossils by radiometric
methods. We have to rely on the evolution of the animals accompanying
the hominids. By this method, it has been established that Australopith-
ecus africanus lived in South Africa between 3 and 2 million years ago.
The Makapansgat Formation appears to be the oldest, and its fossils
would appear to be chronologically very close to the youngest fossils of
Australopithecus afarensis. The Sterkfontein hominids might be around
2.5 million years old, with the Taung Child being the youngest represen-
tative of the species. The environment in which the Sterkfontein austra-
lopithecines lived is interpreted as being forested, though not humid – it
seems more likely that it was dry woodland or scrub with open spaces. In
other words, a mosaic of ecosystems.

Distinguishing Marks

The distinguishing features of our species are a large brain, a unique

capacity to create various kinds of tools using widely varying materials,
articulated language, a long childhood, implying a long learning period,
and a bipedal mode of locomotion (as well as a very specific sexuality,
which we will consider later). The characteristics of large brain, slow
development, and capacity to use or adapt natural objects to make tools
are also found in our closest relatives, the chimpanzees, gorillas, and

Arsuaga / Chosen Species 1405115327-4-004 Final Proof page 63 4.6.2005 2:36pm

The Origin of Humanity

Figure 4.3 Female Australopithecus africanus

Arsuaga / Chosen Species 1405115327-4-004 Final Proof page 64 4.6.2005 2:36pm

The Origin of Humanity

orangutans – naturally at a much lower level of development, but com-

paratively greater than that of other animals. These features, plus the
capacity for language, can be grouped under the label of something we
understand intuitively, but which is impossible to define or measure, and
which we call intelligence. Locomotion is another question, and since
Darwin, science has been asking whether increased intelligence preceded
the erect posture, or vice versa, or whether the two evolved at the same
time. Which is the same as asking: What was the initial impulse in our
evolutionary history – in other words, what made us human?
Stanley Kubrick’s film 2001: A Space Odyssey, based on the book by
Arthur C. Clarke, and hailed as a masterpiece even when it was released in
1968, offered an answer which was very much in line with what was
thought in some scientific circles at the time. The film depicts a group
of apes which can be recognized as hominoids. They are not bipedal, and
they are shown against a background of savanna – presumably in Africa –
very arid, almost desert. These creatures shelter at night in caves to protect
themselves against leopards, and fight over a water pool with a rival group
of apes. In other words, they do not yet show any of our distinguishing
features. Suddenly they find themselves in front of a monolith of extra-
terrestrial origin, which they touch. Then comes the spark which initiates
human evolution: they have an idea. This idea is to use an animal bone as a
tool. For what purpose? In order to kill, in an orgy of blood, first an
animal, and then their enemies in the rival group. In other words, our
ancestors discovered technology and became both carnivorous and killers
of their fellows.
The idea that the first hominids were hunters, or to put it more crudely,
‘‘murdering apes,’’ was developed by Dart during the 1950s. Dart be-
lieved that the australopithecines were hunters and cannibals and, most
importantly, that we have inherited the heavy burden of those violent
instincts, at the same time as improving on their weapons. For the
australopithecines, according to Dart, did not have dressed stone tools;
instead they used weapons made from bones, teeth, and animal horns, an
industry that Dart called ‘‘osteo-donto-keratic,’’ in reference to these
three types of material. According to this theory, it was the hunting
instinct and the taste for flesh that led the early hominids to leave the
trees and make their first weapons, refining their intelligence and also
favoring the adoption of the erect posture, which was certainly more
appropriate for a warrior than the quadruped trot. It is worth noting, in
passing, that in the past, reflections on human evolution frequently sug-
gested that intelligence developed as hominids were confronted with new
challenges – first the savanna and later the colder climates of Europe – but

Arsuaga / Chosen Species 1405115327-4-004 Final Proof page 65 4.6.2005 2:36pm

The Origin of Humanity

stagnated among those who opted for the ‘‘comfort’’ of the forest and its
abundant fruits, or the warm African continent.
Although Dart’s interpretation of our origins was not universally
accepted, he was not alone. In a famous book entitled Man-apes or Ape-
men: The Story of Discoveries in Africa, published in 1967, Sir Wilfrid Le
Gros Clark (1895–1971) opined that the australopithecines were too
poorly defended, with their small canine teeth, to survive without
weapons, whether of stone, bone, horn, or teeth. According to Le Gros
Clark, these hominids were hunters and carrion-eaters cast into a hostile
environment; however, walking on two feet left their hands free to ma-
nipulate tools, and it was this that was the stimulus for the development of
intelligence. Le Gros Clark was a great authority on human evolution, and
it was he who, following a trip to South Africa in 1947 to see the original
fossils, helped to change the opinion of the majority of the scientific
community, allowing the australopithecines to be recognized as primitive
members of our line.
In short, the hunter hypothesis suggests a somewhat bloody beginning
to human evolution, though it is nevertheless a beginning. But was this
how things began? Have we been ‘‘murdering apes’’ and makers of tools
from the start, perhaps even before we became bipedal?
In order to answer this question, we shall address the question of
australopithecine locomotion in the next chapter. Later we shall look at
other forms of hominids, Paranthropus, and the first humans, together
with the first stone tools, and then we will discuss evolutionary changes in
intelligence, diet, growth, and sociability.

Arsuaga / Chosen Species 1405115327-4-005 Final Proof page 66 4.6.2005 2:37pm

The Bipedal Chimpanzee

Moreover, if the individuals I am talking about, moved by the need to

grow higher so as to see all at once far and wide, were forced to hold
themselves upright and acquired from that a constant habit from one
generation to the next, there is no doubt once again that their feet
would have insensibly taken on a shape appropriate for holding them in
an upright position.
Jean Baptiste de Lamarck, Zoological Philosophy

The Great Step

The human being is not the only mammal capable of walking on its hind
limbs. We have already noted that anthropoid apes are in the habit of
maintaining their trunk vertical as they move through the trees hanging
by their arms, or simply when they are sitting. But holding the trunk erect
is only half the battle in achieving erect posture and walking on two legs.
The other half involves aligning the legs with the trunk – in other words,
extending the entire body. The great apes sometimes walk on two legs,
but although they hold their trunk almost vertical, they keep the hip and
knee joints flexed, just as when they walk on four legs. Only we humans
are able to take stable steps without large movements of the trunk, and
long strides when we walk, extending our legs far behind the hip; other
mammals take only small, wobbly steps, with large shifts of the trunk.
Part of the reason for this major difference lies in the pelvis. When we
are standing still, more or less steady, the body is stable and the pelvis is

Arsuaga / Chosen Species 1405115327-4-005 Final Proof page 67 4.6.2005 2:37pm

The Bipedal Chimpanzee

horizontal. However, at the moment when we move a leg forward to take

a step, the weight of the body tends to cause the pelvis to lean over toward
the unsupported side of the body, putting the walker at risk of falling. But
this does not happen, because humans have muscles known as abductors,
which stabilize the pelvis and prevent it collapsing too far toward the
unsupported side (Figure 5.1).

Figure 5.1 Position of the gluteus medius muscle in humans and chimpanzees.
In humans the muscle fibers are oriented laterally, and the muscle therefore acts
as an abductor, balancing the pelvis when the opposite foot moves forward. In
chimpanzees the fibers are oriented toward the back, so that the muscle acts to extend
the joint between the pelvis and femur

Arsuaga / Chosen Species 1405115327-4-005 Final Proof page 68 4.6.2005 2:37pm

The Bipedal Chimpanzee

By contrast chimpanzees, for example, have no mechanism for stabiliz-

ing the pelvis, and in order to avoid falling sideways they have to shift the
whole trunk a long way toward the supporting side, so that their walk
becomes extremely swaying and impractical. On two legs, chimpanzees
walk in a way similar to humans who have suffered paralysis of their
abductor muscles. We therefore have to ask whether nonhuman mammals
do not in fact have pelvic abductor muscles, and when and how these
muscles appeared in human evolution.
Although only bones become fossilized, and flesh never does, paleon-
tologists are able to study the function of muscles of which nothing
remains. This is the province of paleobiomechanics, a branch of functional
paleomorphology which applies the principles of mechanics to the body’s
levers in order to reconstruct the movements which beings of the past
were able to make. The absence of muscles in fossils is not an irredeemable
loss, because in fact there are no muscles which specialize exclusively in
abduction, nor in adduction (the opposite movement), nor in flexion or
extension (the opposite of flexion), nor in rotation.
Muscles simply contract when they receive a neural impulse. That is all.
The effect produced by this contraction on the body’s mechanism de-
pends exclusively on the line of action of the muscle. And to determine the
line of action of a muscle, all we need to know is its two points of contact
with the skeleton, points known as the origin and the insertion of the
muscle. Of course, the actual movement may result from the activity of
several muscles which produce the same action (synergistic muscles), or
different and even opposite actions (antagonistic muscles). But as a whole,
an organism’s capacity for movement can be established in all its com-
plexity if we know the lines of action of all the muscles.
In our species the two pelvic abductor muscles are the gluteus minimus
and, most especially, the gluteus medius. The gluteus maximus, which
forms most of the muscular mass of the buttocks, works to extend the
joint between the pelvis and the femur (naturally there is one each side, as
in all other paired muscles). Its action is to align the trunk with the legs. In
a standing person, the gluteus maximus straightens the trunk. We can also
say that in humans the pelvic extensor muscles are also pelvic stabilizers,
but not transverse stabilizers like the abductors: rather they are antero-
posterior stabilizers (from front to back). In humans the gluteus maximus
is not involved in normal walking on a flat surface, but it comes into action
when we run, jump, or go up a hill or up stairs (Figure 5.2).
In a quadruped the extensors of the pelvis–femur joint perform the
important task of extending the two hind limbs alternately and pushing
the body forward on four feet. Sprinters use the same extension when they

Arsuaga / Chosen Species 1405115327-4-005 Final Proof page 69 4.6.2005 2:37pm

The Bipedal Chimpanzee

Gluteus maximus

Gluteus medius



Figure 5.2 Extensor muscles in chimpanzees and humans. In chimpanzees

the three gluteal muscles act as extensors of the pelvis–femur joint. In humans
only the gluteus maximus does this

push themselves out of the starting blocks, where they position themselves
in a quadruped posture to begin with, with the hip joint highly flexed,
before moving immediately into a bipedal posture.
In apes, as in other mammals, the function of the gluteus medius and
gluteus minimus is different from our own, because the muscles have a
different line of action. In chimpanzees and gorillas the three gluteal
muscles (maximus, medius, and minimus) act as hip extensors, never as
abductors; this means that habitual bipedal walking is not possible, but it
favors quadrupedal locomotion.
We have already noted that the function of a muscle is determined by its
line of action. Why is the line of action of the gluteus medius and minimus
different in humans? The answer, to put it simply, is the orientation of the
bony region in which both muscles originate, the iliac crest of the pelvis.
The pelvis is composed of the two coxal bones, one on either side,
and the sacrum at the back, which in its turn is formed by the sacral
vertebrae (the sacrum is thus part of the spinal column and continues into
the coccyx). The coxal bone is actually the result of the fusion of three
bones during adolescence – the ilium, which forms the upper part of
the coxal bone, the ischium, which forms the lower posterior part, and

Arsuaga / Chosen Species 1405115327-4-005 Final Proof page 70 4.6.2005 2:37pm

The Bipedal Chimpanzee

the pubis, which represents the lower anterior part. These three bones
meet in the acetabulum, where the coxal bone articulates with the head of
the femur (see Figure 10.1).
Of all parts of the postcranial skeleton, i.e., the skeleton excluding the
skull and jawbone, the pelvis is probably that which most distinguishes
humans from apes. This is obviously the result of our particular mode of
walking, since the pelvis of apes is not substantially different from that of
other quadruped mammals. Let us look at the main innovations in the
revolutionary architecture of the human pelvis, and consider their biome-
chanical significance.
In quadrupeds the weight of the trunk is transmitted through the four
limbs, but in humans, because we are bipedal, it is transmitted through
the spinal column down to the sacrum, and from there through the coxal
bones to the heads of the two femurs, continuing down to the feet. In
walking, during the phases when we are supported on only one leg almost
all of the body’s weight is transmitted through one coxal bone (on the
supporting side). In order to reduce the stress on the bar of bone running
from the sacro-iliac joint to the coxo-femoral joint, the two joints have
become much closer than they are in chimpanzees and other quadrupeds.
This biomechanical improvement of the pelvis has, as we shall see later, an
undesirable side effect: it greatly complicates childbirth because it reduces
the size of the bony channel through which the full-term fetus has to pass
in order to be born.
The three gluteal muscles originate in the iliac crest and insert in the
femur. The iliac crest forms the larger part of the iliac bone or ilium, and
in apes consists of a high, narrow plate of bone. In humans it is propor-
tionally shorter, because, as we have already noted, the joints between the
coxal bone and the spine and femur are closer together, but most import-
antly, it is relatively much wider than in other apes.
The other major difference in the human iliac crest is its orientation (see
Figure 10.1). In quadrupeds the surface on which the gluteal muscles
originate faces directly backward. This orientation means that contraction
of the gluteal muscles can only produce extension of the hip joint, because
its line of action is posterior: in other words, if the legs are stable they
draw the pelvis backward, while if the pelvis is fixed it draws the legs
In humans, however, the orientation of the iliac crest has changed
radically, so that the surface where the gluteus medius and minimus
originate faces toward one side rather than backward. The biomechanical
result of this is that when the hip pulls sideways, the contraction of these
two muscles produces the effect of abduction, making bipedal walking

Arsuaga / Chosen Species 1405115327-4-005 Final Proof page 71 4.6.2005 2:37pm

The Bipedal Chimpanzee

possible, as the trunk can be balanced at each stride. The gluteus max-
imus, on the other hand, which originates in the hindmost part of the
ilium and in the sacrum, continues to act as an extensor in humans, since
its line of action is posterior rather than lateral.
One of the biggest questions of evolutionary biology is how the great
anatomical transformations which create organisms radically different
from their predecessors come about. A bipedal primate represents a
revolutionary change, not simply a slight variation on other types of
hominoids. We have to rule out the idea that the entire skeleton altered
drastically all at once, but it is not easy to imagine how an organism can
move gradually from walking on four legs to walking on two. One inter-
esting hypothesis is that the initial modification which made it possible for
primates to begin walking bipedally was a change in the orientation of the
iliac crest. Merely altering in this orientation, causing the iliac crest to face
more to the side, would generate some capacity for abduction, one of the
basic features of bipedalism. If walking on two legs increased the possi-
bilities for surviving and reproducing, other modifications would subse-
quently continue to be selected until the entire skeleton was affected.
We will not exhaust the reader with any more biomechanical explan-
ations, although this is a fascinating discipline which allows the paleon-
tologist to play God for a moment, and say to a fossil skeleton: ‘‘Get up
and walk!’’ Just one final detail on the pelvis: in humans the capacity for
extension of the hip, which is very useful for climbing trees and for
walking on four legs on the ground, is also reduced in the hamstring
muscles, which originate in the lower part of the pelvis (Figure 5.3).
Let us turn now to the fossils. So far we have been engaging in an
exercise of comparative anatomy, the best starting point for paleonto-
logical analysis. We have a limited sample of fossil pelvises from australo-
pithecines. The most complete is that of Lucy, which has retained the
sacrum and all of the left coxal bone, and those of two Sterkfontein
skeletons, Sts 14 and Stw 431 (the latter is fairly fragmented and has
not yet been analyzed). Although they show some distinctive character-
istics, their morphology is entirely human and does not resemble that of
apes in any way. The ilium is short and wide (favoring abduction), and the
ischium is short (reduced capacity for extension). Furthermore, the sac-
rum in these pelvises is proportionately wide, not narrow as in apes.
Various authors offer different interpretations of the exact orientation of
the iliac crest, which has had to be reconstructed in Sts 14 and Lucy
because both had been distorted by being crushed in the deposits. Some
authors suggest that the iliac crest in these fossils faces less to the side
and more toward the back than in modern humans, but all researchers

Arsuaga / Chosen Species 1405115327-4-005 Final Proof page 72 4.6.2005 2:37pm

The Bipedal Chimpanzee

Figure 5.3 Hamstring muscles in humans, chimpanzees, and gorillas. These

muscles, consisting of the biceps femoris, semimebranosus, and semitendinosus,
make up the muscular mass of the posterior face of the thigh; they are known
collectively as the hamstrings, and they run from the lowermost part of the
ischium, or ischial tuberosity (the bones on which we sit), to the tibia and fibula.
In all cases their action is to extend the joint between the pelvis and femur

Arsuaga / Chosen Species 1405115327-4-005 Final Proof page 73 4.6.2005 2:37pm

The Bipedal Chimpanzee

recognize that australopithecines had the capacity for abduction of the

pelvis which makes it possible to walk on two legs.
There are other skeletal indications that the australopithecines were
bipedal. In a human standing stably, the diaphysis (the shaft or main
axis) of the femur runs at an angle from the hip to the knee; the knees
are very close together (Figure 5.1). If the knees and the feet, which are
located vertically beneath the knees, were far apart the center of gravity
would have to shift far to the supporting side of the body on each step,
and walking would be less efficient and would use more energy. This
modern morphology of the femur is also found in australopithecines.
However in chimpanzees, for example, the femurs are not inclined down-
ward and inward (quadrupeds do not bring their knees together below
their belly, as they have no problems balancing when they walk); this,
combined with the lack of abductors, means that when walking on two
legs they have to tip the trunk markedly over the supported side of the
body with each step, to ensure that the center of gravity does not project
beyond the supporting foot. Thus their walking is very inefficient and
energy-intensive, as it requires large movements of the body in order to
move forward a very small distance.
Another classic feature indicating that australopithecines were bipedal is
the position of the foramen magnum, the opening at the base of the skull
through which the spinal cord runs. In humans the foramen magnum is
oriented downward because the spine is positioned vertically when we
walk. This does not mean that the spine runs in a straight line: in fact in
humans (as perhaps in australopithecines too) the spine is markedly
curved, with a cervical and a lumbar lordosis (curves toward the front in
the regions of the neck and lower back, respectively), and a thoracic
kyphosis (curve toward the back in the region of the ribs) (Figure 5.4).
In apes the foramen magnum is further behind the base of the skull and
is oriented more toward the back, because in quadrupedal walking the
head is positioned on the end of a diagonal spine (with no cervical or
lumbar curves). The nuchal plane, i.e., the part of the occipital bone in
which the muscles of the back of the neck which hold the head in position
originate, is also wider and oriented more toward the back in apes than in
humans. Australopithecus afarensis appears to be at an intermediate point
between these features, leading some authors to reconstruct the neck of
this species as more inclined than ours, but more vertical than that of
However, not all the experts agree on how the australopithecines
walked on two legs. Some, among which we number ourselves, believe
that their way of walking was not substantially different from our own,

Arsuaga / Chosen Species 1405115327-4-005 Final Proof page 74 4.6.2005 2:37pm

The Bipedal Chimpanzee

Cervical curve
Sagittal ridge (lordosis)
Nuchal ridge

Orientation of the
Orientation of the
foramen magnum
foramen magnum
Thoracic curve

Lumbar curve

Figure 5.4 Curves of the spine and orientation of the foramen magnum in
gorillas and humans

while others believe that it was less ‘‘perfect.’’ The discussion might
have continued ad infinitum had not the team directed by Mary Leakey
(1913–96) made a completely unexpected discovery in 1978 and 1979:
several meters of tracks formed by footprints left by three hominids
who walked through an area of what is now Tanzania about 3.5 million
years ago.

The Laetoli Footprints

The Laetoli Beds are located not far from the famous Serengeti National
Park. During one of its eruptions a nearby volcano, Sadiman, threw ash
out into the air, and rain converted this to mud in which the tracks of
many animals were recorded and fossilized. Among these animals were the
hominids we have mentioned. There are two parallel tracks, but that on
the right (looking in the direction of walking) appears to be that of two

Arsuaga / Chosen Species 1405115327-4-005 Final Proof page 75 4.6.2005 2:37pm

The Bipedal Chimpanzee

individuals, one of them walking in the tracks of the other. The individual
on the left is very small, perhaps a female or a child.
The characteristics of these footprints are incredibly modern. They
show no sign of insecure or ‘‘imperfect’’ bipedalism; rather they indicate,
even in the smallest details, a way of walking identical to our own. The
foot of a chimpanzee, gorilla, or orangutan is very different from a human
foot. In fact it is more like our hand – flat, with a big toe which is shorter
than the others and can separate from them laterally, and extends away
from the other toes on each step (Figure 5.5).
In the Laetoli footprints we can study the mode of walking of the
hominids which produced them. If you are fortunate enough to have a
beach nearby, you can compare these fossil footprints with your own, and
appreciate their extraordinary similarity. In each step taken by the Laetoli
hominids the front foot was supported first on the heel, leaving a deep
impression in the soft ground. Part of the body weight was then trans-
ferred through the arch or instep. After this, the foot flexed over the toes,
giving a final impulse to lift the foot from the ground and extend the leg
forward. As in modern humans, the big toe was fundamental to this final
phase (being the last to leave the ground), and it therefore pointed
forward like the other toes and was aligned with them (Figure 5.5).
Fossils of Australopithecus afarensis of the same age as the footprints
have been found in the Laetoli Beds, and we can therefore only suppose

Figure 5.5 Footprints of a chimpanzee (left), a modern human (center), and

a Laetoli hominid (right)

Arsuaga / Chosen Species 1405115327-4-005 Final Proof page 76 4.6.2005 2:37pm

The Bipedal Chimpanzee

that the footprints were produced by three individuals of this species of

However, the researchers who, on the basis of analysis of fossil bones,
had concluded that individuals such as Lucy could not have had a bipedal
locomotion so similar to our own (including Brigitte Senut, Randall Sus-
man, Jack Stern, William Jungers, Russell Tuttle, and Peter Schmid),
remain unconvinced by the evidence of the Laetoli footprints. Their
explanation is that 3.5 million years ago there were two radically different
types of hominids in East Africa. On the one hand there were the hom-
inids like Lucy, who were bipedal but had a mode of walking that was not
as fully developed as that of humans, and spent much of their time in the
trees, where they felt more comfortable; these hominids were close rela-
tions, but not our ancestors. Then there were some other, enigmatic,
hominids, our direct ancestors, who walked on two feet in the modern
way, and it was they who created the Laetoli footprints.
We have already made clear our position on this question. In our
opinion, and that of many other authors, the fossils from Laetoli in
Tanzania and from Hadar in Ethiopia represent a single species, Australo-
pithecus afarensis, which varied widely in size but had a single mode of
locomotion. Until new fossils indicating the contrary appear, these aus-
tralopithecines are the only candidates for having made the Laetoli foot-
prints. In other words, Lucy walked like us.

The Mystery of Mysteries of Human Evolution

Having discussed the origin of one of our principal distinguishing fea-

tures, the erect posture, there remains one basic and inevitable question:
what is it for? The traditional answer was that walking on two legs was an
adaptation to the savanna, in order to see over tall grasses or something
along those lines. However, there is a conceptual error here. Animals
adapt not to environments, such as savanna, forest, or sea, but to eco-
logical niches, to the roles that species play in ecosystems, which may be
very varied. To put it another way, there are many species in the modern
savanna and none of them is bipedal except for ours. So what we need to
ask is what kind of ecological niche the first hominids to become bipedal
were occupying. Moreover, it now seems that australopithecines were
forest-dwellers rather than inhabiting open environments.
Before returning to this question it is worth pausing for a moment
to rehabilitate the erect posture, which is traditionally denigrated and

Arsuaga / Chosen Species 1405115327-4-005 Final Proof page 77 4.6.2005 2:37pm

The Bipedal Chimpanzee

considered inefficient, an evolutionary botch job whose enormous draw-

backs had to be compensated by some major advantage. In general it was
thought that the advantage was that the hands were freed from walking,
allowing for the manufacture of tools and the development of the brain.
However, as we shall see, all of these things came long after we became
Bipedalism involves an extensive reorganization of the skeleton, and this
was achieved with remarkable perfection from the engineering point of
view. We can see our body is as if it were built from Meccano: it is made up of
a number of articulated segments whose individual centers of gravity are
located in the same plane as the axes of the main articulations between the
segments (hips, shoulders, elbows, wrists, knees, ankles, etc.). This is at the
same time the plane which contains the center of gravity of the entire body.
This means that our standing posture is very stable, and requires virtually
no effort to maintain. Only the center of gravity of the head is slightly
forward of its joint with the first cervical vertebra. This disadvantage has to
be compensated by the muscles of the back of the neck, which hold the
head erect. However, the reduction of the facial skeleton over the course of
human evolution has resulted in a considerable improvement in this prob-
lem, shifting the head’s center of gravity backward.
One way of assessing the biomechanical efficiency of a way of moving is
to follow the trajectory of the body’s center of gravity. In our case, the
center of gravity is located in front of the second sacral vertebra, more or
less at the level of the navel. If we watch a person walking from the side,
we will see slight rising and falling movements of the head, corresponding
to rises and falls in the center of gravity. Seen from the front, the walker’s
head inclines slightly toward the supported side at each step.
The straighter the trajectory of the center of gravity, the more econom-
ical, in terms of energy consumption, is the walk. If the center of gravity
travels on a very winding path, with marked rises and falls and large lateral
shifts, the walk will be inefficient and will waste energy. This is what
happens when chimpanzees walk on two legs. However, in humans the
center of gravity travels on an almost straight line in walking.
So in physical terms our way of walking is no less successful than that of
quadrupeds, although of course it is not as fast over short distances. On
the other hand we have great stamina, better than that of many quadru-
peds, when we need to travel long distances or over long periods of time,
whether running or walking. And in fact chimpanzees’ and gorillas’ way of
walking, supporting themselves on the middle phalanges of the hands,
cannot be considered a marvel of adaptation to life on land; rather it seems
to be a compromise solution to the problem of needing to move both

Arsuaga / Chosen Species 1405115327-4-005 Final Proof page 78 4.6.2005 2:37pm

The Bipedal Chimpanzee

through the trees and over the ground. It appears that gorillas later largely
abandoned life in the trees when they became very heavy.
Peter Wheeler has identified a further advantage in the shift of the body
to the vertical, related to body temperature. An individual standing on
two feet receives less solar radiation, especially when the sun is at its
highest point, than a quadruped. Moreover, when the body leaves the
ground it moves away from a focus of heat and benefits from cooling
breezes. Combining this advantage with that we noted above, we might
conclude that bipedal locomotion is perhaps the best solution for a
hominid who has to travel long distances while exposed to the sun. The
first bipedal hominids did not live in the savanna, but they might never-
theless have had to travel between patches of vegetation separated by open
One indisputable benefit of bipedalism is the ability to use the hands
and arms for carrying, whether food or perhaps children. In this, quad-
rupeds are at a clear disadvantage to us. We shall return to this capacity
for carrying in the chapter on the social biology of the first hominids,
because one proposed explanation for the origin of bipedalism is partly
based on this factor. Without jumping ahead, we can already anticipate
that such an explanation will have many problems, in terms of what we
intuit about the social biology of the first hominids. Thus we can still
today consider the significance of our acquisition of our special way of
traveling as one of, if not the biggest problem confronting human

Portrait of the Entire Body of an Australopithecine

Lucy was a very small individual. She was about 105 cm tall and weighed
no more than 30 kg. But she was not exceptional among her species. Even
the first fossils discovered by Donald Johanson in 1973, consisting of a
knee joint (the lower end of the femur and the upper end of the tibia),
correspond to an individual of the same size as Lucy; some of the hom-
inids from site 333 are of similar size (as we noted above, this assemblage
of fossils was christened the First Family). Other fossils are larger and
correspond to individuals around 135 cm tall, weighing about 45 kg. It is
thought that the small individuals, like Lucy, were females, and the larger
ones males. The male and female averages may have been somewhat closer
in Australopithecus africanus, but in any case australopithecines were small
hominids, similar in size to chimpanzees.

Arsuaga / Chosen Species 1405115327-4-005 Final Proof page 79 4.6.2005 2:37pm

The Bipedal Chimpanzee

In addition to being small, the proportions of Lucy’s limbs were dif-

ferent from ours, including those of modern populations of small stature
such as the Pygmies. The most distinctive feature of Lucy’s postcranial
skeleton is the shortness of the legs (Figure 5.6). The ratio of the length of
the humerus to that of the femur (which expresses the relation of arm
length to leg length) is 85 percent, markedly higher than that of humans,
which averages 71 percent, and markedly lower than that of orangutans
(129 percent) and gorillas (118 percent), but not so different from
common chimpanzees (102 percent) and bonobos (98 percent). Chim-
panzees have relatively longer forearms than we do: the ratio between the
ulna (one of the bones of the forearm) and the humerus is around 80
percent in humans, and around 95 percent in chimpanzees (even higher in
orangutans and gibbons). In Lucy the ratio is estimated at 92.5 percent.
These proportions suggest that the australopithecines still maintained
some of their ancestors’ aptitude for climbing trees. Moreover, the phal-
anges of the fingers and particularly of the toes of Australopithecus afar-
ensis are curved, resembling those of chimpanzees more than any other
hominid; this curvature of the phalanges relates to the capacity for grasp-
ing branches and moving through the trees. All of these factors, combined
with the more or less enclosed forest habitat in which they are thought to
have lived, leads us to believe that in addition to being able to walk on two
legs, australopithecines still climbed trees to find food, to escape pred-
ators, or to sleep. Chimpanzees use branches and leaves to construct a
kind of personal nest in the tops of trees where they can spend the night.
Gorillas also do this, although on the ground, because of their weight. It
would not be surprising if the australopithecines still retained this habit of
sleeping in the trees (see Figure 11.2).
Despite the extraordinary amount of information which Lucy and other
fossils give us, there are still many features of the australopithecine body as
a whole which remain unclear. We have already noted that in the great
apes the lumbar section of the spine is proportionally shorter. This means
that the last ribs are very close to the pelvis. The great apes therefore lack
our characteristic narrowing of the waist. Furthermore, in chimpanzees,
gorillas, and orangutans the rib cage widens toward the base, taking the
form of a cone (see Figure 2.4).
In humans the lumbar section has become a little longer, and as the
ilium has also become much shorter we have a capacity to rotate the
thorax (ribcage) at this level. This is very useful to us in bipedal walking,
since as we bring the hip forward on the side of the front foot, the
shoulders turn in the opposite direction. Moreover, our ribcage is bar-
rel-shaped (Figure 2.4).

Arsuaga / Chosen Species 1405115327-4-005 Final Proof page 80 4.6.2005 2:37pm

The Bipedal Chimpanzee

Figure 5.6 Side view of a male skeleton of Australopithecus afarensis

Arsuaga / Chosen Species 1405115327-4-005 Final Proof page 81 4.6.2005 2:37pm

The Bipedal Chimpanzee

In australopithecines the marked shortening of the ilium seems to have

clearly separated it from the lowest ribs (Figures 5.7–5.10). Furthermore,
there were more lumbar vertebrae (five or six) than in the great apes
(which have three or four). All of this meant a greater capacity for turning
at the waist. However, the shape of the rib cage might have been more like
that of the great apes.
To complete the full portrait of the australopithecine body, we still need
the head. If we want to picture one of these small bipedal hominids, it is
best to start by giving it the head of a chimpanzee or a female gorilla.
From here we have to shade in the details. The skull is divided into two
parts – the facial skull, the skeleton of the face, and the cerebral skull or
neurocranium, in which the brain is encased. The australopithecines’ brain
was little larger than that of chimpanzees, and therefore the neurocranium
would not be much bigger. There are, however, some differences. We
have already noted the position of the foramen magnum, which is related
to posture and distinguishes the australopithecines from all the apes,
although the neck was somewhat inclined and the head further forward
in the first hominids (Figure 5.6).
In male orangutans and gorillas, the temporal muscles (which are in-
volved in chewing, and of which we will speak more later) are so highly
developed that they need a greater bony surface than is provided by the
walls of the neurocranium, and thus a bony ridge forms over the whole of
the top of the cranium, along the middle or sagittal plane (Figure 5.4). At
the back this sagittal ridge merges with another, transverse ridge which
forms the upper limit of the nuchal plane and results from similarly highly
developed muscles in the back of the neck. These ridges and the muscles
originating in them give the heads of male gorillas, the old silverbacks, a
pointed appearance, which the reader has perhaps noted. Sagittal and
nuchal ridges are not frequently found in male chimpanzees, but they are
found, although in more attenuated form, in some examples of Australo-
pithecus afarensis which have been judged to be male, such as A.L. 444-2
(Figure 4.2). It is not clear whether male Australopithecus africanus had
similar sagittal ridges. Another feature distinguishing australopithecines
from apes is the brow ridge. This is a bony rim above the eye socket,
which in chimpanzees and gorillas is clearly marked and separated from
the rest of the frontal bone by a furrow, to the point where it can clearly be
seen in a living animal (see Figure 2.2). Australopithecines do not have a
brow ridge as such: the bony rim is not separated from the frontal bone by a
furrow (compare Figures 4.2 and 4.3 with Figure 7.4).
The facial skeleton of australopithecines appears little different from
that of chimpanzees and gorillas. The face of australopithecines projected

Arsuaga / Chosen Species 1405115327-4-005 Final Proof page 82 4.6.2005 2:37pm

The Bipedal Chimpanzee

Figure 5.7 Front view of the skeleton of a male Australopithecus afarensis.

The rib cage is reconstructed as funnel-shaped, as in apes

Arsuaga / Chosen Species 1405115327-4-005 Final Proof page 83 4.6.2005 2:37pm

The Bipedal Chimpanzee

Figure 5.8 Reconstruction of the musculature of a male Australopithecus


Arsuaga / Chosen Species 1405115327-4-005 Final Proof page 84 4.6.2005 2:37pm

The Bipedal Chimpanzee

Figure 5.9 Impression of a male Australopithecus afarensis

Arsuaga / Chosen Species 1405115327-4-005 Final Proof page 85 4.6.2005 2:37pm

The Bipedal Chimpanzee

Figure 5.10 Impression of a female Australopithecus afarensis

Arsuaga / Chosen Species 1405115327-4-005 Final Proof page 86 4.6.2005 2:37pm

The Bipedal Chimpanzee

well forward, showing a high degree of prognathism, to use the technical

term: the nose did not stand out from the face because the nasal bones did
not form a bony bridge.
There are of course many other distinguishing features which are
of interest to the specialist, but basically, the head of an australopithecine
is thought to have been generally similar to that of an African great ape
of the gorilla or chimpanzee type, but with its own particular features –
among which the small canine teeth are perhaps the most notable.

Arsuaga / Chosen Species 1405115327-4-006 Final Proof page 87 4.6.2005 2:38pm

Paranthropus – Hominids of
the Open Plains

Gert was found, and drew from his trouser pocket four of the most
wonderful teeth ever seen in the world’s history.
Robert Broom, The South African Fossil Ape-Men

The Emergence and Distribution of Paranthropus

As we have already noted in a previous chapter, marked climatic fluctu-

ations, associated with the expansion of the ice sheets in the northern
hemisphere during the cold eras, began to occur about 2.8 million years
ago. This change in climate had important consequences for the flora and
fauna of the East African equatorial regions, where grassland spread to
replace other, more wooded environments. The small mammals and
bovids (especially the antelopes) found in faunas of this era provide a
record of how forms typical of humid forest environments were replaced
by other species more adapted to open, arid environments; these are still
found in the savanna today.
On the basis of these data, Elisabeth Vrba has suggested that this climatic
and ecological crisis could have had a decisive influence on the evolutionary
history of hominids, contributing to the disappearance of Australopithecus
africanus (still linked to forest environments), and favoring the selection
of new forms adapted to life in more open environments – the first
representatives of the genera Paranthropus and Homo (humans).
The term Paranthropus literally means ‘‘beside man,’’ and was coined
by Robert Broom (1866–1951) in 1938 to name a series of hominid

Konso Paranthropus boisei

Omo Paranthropus aethiopicus

Paranthropus boisei

Lake Turkana Paranthropus aethiopicus

Paranthropus boisei

Olduvai Paranthropus boisei

Arsuaga / Chosen Species 1405115327-4-006 Final Proof

Lake Malawi Paranthropus

page 88

Swartkrans Paranthropus robustus

4.6.2005 2:38pm

Kromdraai Paranthropus robustus

Figure 6.1 Location of the main deposits containing fossils of Paranthropus

Arsuaga / Chosen Species 1405115327-4-006 Final Proof page 89 4.6.2005 2:38pm

Paranthropus – Hominids of the Open Plains

fossils he had found in the South African Kromdraai deposit. The term is
very appropriate, since Homo and Paranthropus emerged during the same
era and lived alongside one another for about 1.5 million years, inhabiting
a vast region extending from present-day Ethiopia to the southern tip
of Africa.
The majority of paleontologists believe that there were three species
within the genus Paranthropus. Two of these are exclusively East African:
Paranthropus aethiopicus (aethiopicus refers to Ethiopia, the country
where the first remains of this species were found) and Paranthropus boisei
(the term boisei is in honor of an English sponsor named Boise); the third
has been found only in deposits in the south of the continent, and is
known as Paranthropus robustus (robustus meaning ‘‘robust’’ in Latin).
Although the first remains attributed to the genus Paranthropus –
specifically to Paranthropus robustus – appeared in South Africa, the oldest
known fossils were found thousands of kilometers away. In 1967, Camille
Arambourg (1885–1969) and Yves Coppens published details of a hom-
inid jawbone (Omo 18-1967-18) found in one of the deposits of the
lower reaches of the Omo River (close to where it flows into Lake
Turkana), in Ethiopia; the age of these fossils was established as 2.6
million years. This discovery passed almost unnoticed until, in 1986,
Richard Leakey and his collaborators published the find of an extraordin-
ary skull (WT 17000) belonging to the same species as the jawbone found
by Arambourg and Coppens two decades before, in one of the deposits on
the eastern shore of Lake Turkana. This fossil is one of the amazing
‘‘missing links’’ which paleontologists are occasionally lucky enough to
come upon, since it documents an evolutionary stage intermediate be-
tween Australopithecus afarensis and Paranthropus boisei.
Dated at around 2.5 million years old, WT 17000 is the most repre-
sentative fossil of the species Paranthropus aethiopicus (Figure 6.2). Many
of the features typical of the anatomy of Paranthropus are not yet present
in this species, while others are only in the incipient stages of develop-
ment. The most recent fossils attributed to Paranthropus aethiopicus are a
series of remains of skulls and mandibles found in the river Omo deposits,
and dated at around 2.3 million years old.
The first fossil of Paranthropus boisei was found in 1959, in Olduvai
Gorge, Tanzania, by Mary Leakey and Louis Leakey (1903–72), Richard
Leakey’s parents. It consists of an almost complete skull which its finders
dubbed ‘‘Dear Boy,’’ since it belonged to a young specimen. The age of
this fossil (known as OH5) is around 1.8 million years, and the Leakeys
attributed it to a new species and genus, Zinjanthropus boisei, although it
came to be included in the Paranthropus genus soon afterward.

Arsuaga / Chosen Species 1405115327-4-006 Final Proof page 90 4.6.2005 2:38pm

Paranthropus – Hominids of the Open Plains

Sagittal ridge


ramus Mandibular


Figure 6.2 Male Paranthropus aethiopicus

Arsuaga / Chosen Species 1405115327-4-006 Final Proof page 91 4.6.2005 2:38pm

Paranthropus – Hominids of the Open Plains

Paranthropus boisei is now a relatively well-known species (Figure 6.3),

whose oldest fossils consist of a fragment of jawbone and some teeth
about 2.3 million years old, found in the deposits of the Omo River valley.
The youngest known fossils of this species (two milk teeth), which are
about one million years old, also come from these deposits.
The most extensive collection of fossils of Paranthropus boisei has come
from deposits on the eastern and western shores of Lake Turkana, lying
within the Koobi Fora stratigraphic formation, and has been built up by
Richard Leakey’s team from the late 1970s onwards. In addition to large
numbers of teeth and jawbones, some skulls have been found, and among
these that known as ER 406, ascribed to an adult male, stands out.
A partial skeleton (ER 1500), attributed to an adult female, has also
been found at Koobi Fora. On the basis of these and other remains, it
has been established that the height, weight, and body proportions of
Paranthropus boisei were basically similar to those of the australopithec-
ines (Figure 3.4).
In addition to those at Olduvai, the Omo River valley, and the shores of
Lake Turkana, other East African deposits between 1.3 and 1.4 million
years old have yielded fossils of Paranthropus boisei. In 1964 one of the
most complete mandibles known of this species was found in the Peninj
deposits in Tanzania (Peninj 1); in 1970 part of a skull (known as CH 1)
was found at the Kenyan site of Chesowanja. In October 1997 a team led
by Gen Suwa published the find, in the Ethiopian Konso deposit, of the
only skull complete with mandible known for this species (KGA 10-525);
it was dated at 1.4 million years old.
While the discovery of a skull together with its jawbone is very import-
ant, there is another aspect of the Konso find which is especially signifi-
cant. The bovid fauna in this deposit consists of 80 percent alcelaphins
(animals of the wildebeest type, such as damaliscus and topi), while fossils
of equines are also frequent. Alcelaphins are a group (technically a tribe) of
bovids which eat C4 plants (as do equines); as we have already seen, these
form the predominant vegetation in grasslands. In other words, the fossil
fauna indicates an open, dry environment of the savanna type.
As we have already noted, Paranthropus fossils have also been found in
the far south of the continent; these are assigned to the species Paran-
thropus robustus. Broom’s first finds in Kromdraai were supplemented by
others made by Broom together with John Robinson ten years later, in
1948, in the neighboring Swartkrans cave.
On the basis of paleontological analysis, the stratigraphic levels contain-
ing fossil Paranthropus in Kromdraai and Swartkrans are dated between
1.8 (the oldest) and 1 million years (the most recent) old.

Arsuaga / Chosen Species 1405115327-4-006 Final Proof page 92 4.6.2005 2:38pm

Paranthropus – Hominids of the Open Plains

Figure 6.3 Male Paranthropus boisei

Arsuaga / Chosen Species 1405115327-4-006 Final Proof page 93 4.6.2005 2:38pm

Paranthropus – Hominids of the Open Plains


Zygomatic bone

Zygomatic arch




Zygomatic bone

Zygomatic arch


Figure 6.4 Bones and muscles involved in mastication. The temporalis muscle
is fan-shaped. Its wider end originates on the external surface of the side walls of the
cranium, while the narrower part passes through the temporal fossa to insert in the
mandible. Contraction of these muscles can easily be felt by palpating the temples
while clenching the teeth. The masseter muscles extend from the lower edge of
each zygomatic arch to the outer faces of the mandibular ramus and the posterior
parts of the mandibular body (see Figure 6.1). It is also easy to locate these
muscles by touching the cheekbones while clenching the molars

Arsuaga / Chosen Species 1405115327-4-006 Final Proof page 94 4.6.2005 2:38pm

Paranthropus – Hominids of the Open Plains

Most of the fossils found in Kromdraai are fragments of skulls, jawbones,

teeth, and a few fragments of arm, pelvis, and foot bones. The collection
from Swartkrans is larger, and includes a fairly complete, though distorted
skull (SK 48) as well as other skull and jawbone fragments and teeth. There
are also numerous postcranial fragments, including arm, leg, and pelvis
bones, and vertebrae. Nevertheless, the attribution of these postcranial
remains to Paranthropus is problematic, given that fossils of a primitive
representative of the genus Homo are found at the same levels.
To complete this account of the geographic distribution of Paranthro-
pus, we should note that the great fossil ‘‘gap’’ between Tanzania and
South Africa might begin to be filled with the finds now being made by a
team led by Friedemann Schrenk and Tim Bromage, in deposits on the
eastern shore of Lake Malawi (in Malawi). In addition to a jawbone
attributed to the genus Homo, which we shall describe later, it is known
that this team has found a fragmented upper jawbone of Paranthropus,
details of which have yet to be published.

The Specialist

To the biologist, a specialist organism is one that has adapted to a very

specific lifestyle (or ecological niche). This specialization will usually affect
its diet, which in turn will be reflected in marked adaptations of its
chewing and digestive apparatus. Conversely, organisms which maintain
a more varied lifestyle and diet, and are thus less specialized, are known as
Specimens of Paranthropus show a series of unique traits which allow us
to distinguish them easily from other hominids. These characteristics
relate to their impressive masticatory (chewing) apparatus, which suggests
marked specialization in diet among Paranthropus.
Mastication, or chewing, is the process whereby food is crushed by the
pressure of teeth set in a mobile bone (the mandible or lower jawbone)
against teeth set in a fixed bone (the maxilla or upper jawbone). As in a
mortar, the food can be simply crushed, if the mandible moves only up
and down, or ground, if this vertical movement is accompanied by other,
horizontal movements. Both the pressure of the mandible on the maxilla
(i.e., the strength of the bite) and the direction of the movements made
depend on the muscles attached to the mandible.
The movements of the mandible during chewing are very complex, and
involve many muscles. However, we can simplify the effects which interest

Arsuaga / Chosen Species 1405115327-4-006 Final Proof page 95 4.6.2005 2:38pm

Paranthropus – Hominids of the Open Plains

us, noting that vertical movements of the mandible are generated basically
by two groups of muscles located symmetrically at either side of the head:
the temporalis muscles and the masseter muscles (Figure 6.4). The hori-
zontal movements of the mandible, on the other hand, depend principally
on the action of the lateral pterygoid muscles.
The three species of Paranthropus show different degrees of specializa-
tion of the masticatory apparatus, which is very marked in Paranthropus
boisei, less pronounced in Paranthropus robustus, and only sketchily ap-
parent in Paranthropus aethiopicus.
Beginning with dentition: the molars of Paranthropus have a fairly thick
layer of enamel, thicker than that of any other hominid or ape. This
characteristic is clearly related to a diet which causes heavy wear on the
molars, either because the food contains hard particles or because it
requires prolonged chewing, or both. Moreover, the surface of the teeth
used to crush food is very large in relation to body weight. This is due to
the increased size of the molars and particularly to the fact that the
premolars are ‘‘molariform’’ – they are shaped like molars, substantially
increasing the total area of teeth dedicated to crushing food. Parallel with
the increase in the size of the molars and premolars, the anterior teeth of
Paranthropus, the canines and incisors, are much reduced in size.
In apes the premolars are not ‘‘molariform,’’ so the area of the teeth
dedicated to crushing food is restricted to the molars (and the reader will
recall that the incisors and canines are large). In Paranthropus, in contrast,
the mastication zone has extended to the premolars, which in these species
are ‘‘molariform.’’
The maxilla (upper jawbone) of Paranthropus is further back under the
cranium than is usually seen in apes and australopithecines, and
the premolar area is thus also further back. The zygomatic bones, on the
other hand, are further forward, bringing the zone in which the masseter
muscles originate further forward. These modifications increase the
strength of the bite in the premolar area (Figure 6.4).
Furthermore, the force exerted by a muscle is directly related to the
number of muscle fibers it contains – in other words, to its thickness.
There is a physical limit on the size of the temporalis muscle: it has to pass
through the temporal fossa, the diameter of which determines the max-
imum thickness of the muscle. In Paranthropus the zygomatic arch (which
forms the lateral end of each of the temporal fossae) is strongly curved
outward, markedly increasing the diameter of each fossa and indicating
extremely thick and powerful temporalis muscles. These muscles were so
big in Paranthropus that in males they result in the appearance of a sagittal
ridge along the mid-line of the upper surface of the cranium.

Arsuaga / Chosen Species 1405115327-4-006 Final Proof page 96 4.6.2005 2:38pm

Paranthropus – Hominids of the Open Plains

In addition to the increase in size of the temporalis muscle, the man-

dibular ramus on each side is very long, increasing the power of the action
of the temporalis and masseter muscles. This results in a corresponding
increase in the height of the zygomatic bones and the maxilla.
The mandible (lower jawbone) in Paranthropus is very robust (that is, it
has a very thick mandibular body), responding to the need for the bone to
resist and dissipate the powerful stresses caused by the great force of
chewing. In addition, a Paranthropus mandible is easily distinguished
from that of any other hominid or ape because it is very wide in relation
to its length. This form is ideal for circular movements of the mandible,
used to grind hard food.
The effect of these skeletal modifications on the physiognomy (facial
structure) of Paranthropus was considerable (Figure 6.3). As a result of
the backward shift of the maxilla and the forward shift of the cheekbones,
the face of Paranthropus was not prognathic, with the face projected
forward, but flat or even concave. Moreover, the region of the cheekbones
was much wider because of the lateral widening of the temporal fossae.
Finally, the face was also very long, because of the increased length of the
mandibular ramus and the zygomatic bones and maxilla. It is the unmis-
takable face of very specialized hominids.

Arsuaga / Chosen Species 1405115327-4-007 Final Proof page 97 4.6.2005 2:38pm

A New Kind of Hominid

Whatever a man comes to be worth, he can never hope to exceed his

innate worth as a man.
Antonio Machado, Juan de Mairena

The First Humans

The term homo (‘‘man,’’ in the generic sense of ‘‘human being’’) was used
by Linnaeus in 1758 to denote the genus to which our own species (Homo
sapiens) belongs. The terms ‘‘humanity’’ and ‘‘human’’ are usually
restricted to representatives of our genus, so when we explore the origin
and evolution of the genus Homo we are referring to the origin and
evolution of humans.
Until the early 1990s the oldest fossils assigned to our genus consisted
of an assemblage of isolated teeth (around 2.1 million years old) and
jawbone fragments (around 2.5 million years old) found in the deposits
of the Omo River valley. However, the allocation of these fossils to the
genus Homo was disputed by many authors, who called for more solid
evidence that Homo did in fact exist more than 2 million years ago. Those
who believed that the genus was in existence at this time based their
argument on the identification of stone tools in sediments about 2.3
million years old, also in the Omo deposits. But this was a debatable
argument based on a premise which had yet to be proved: that the only
hominids capable of manufacturing stone tools are members of the genus

Arsuaga / Chosen Species 1405115327-4-007 Final Proof page 98 4.6.2005 2:38pm

A New Kind of Hominid

Since the early 1990s, a series of fossils over 2 million years old, and
attributed to our genus, has been presented to the scientific community.
In 1992 it was reported that a fragment of temporal bone (known as
BCI, and found in 1965 in a deposit close to Lake Baringo in Kenya)
belonged to our own genus; this was dated at around 2.5 million
years old. However, as we do not know exactly where it was found,
its position in the stratigraphy of the deposit cannot be determined, and
its age cannot be reliably established. Moreover, not all researchers
accept that this fossil belongs to the genus Homo. In our view, the
Lake Baringo fossil does belong to our genus, although it cannot be
assigned to a particular species. A very complete mandible (UR 501),
found by Schrenk and Bromage’s team on the western shore of Lake
Malawi, is also attributed to Homo (specifically, to Homo rudolfensis).
This fossil is dated at between 2.3 and 2.5 million years old, on the basis
of paleontological data; this age is considered approximate and could be
reappraised. Again, the attribution of this mandible to Homo cannot
be considered definitive, given that, as we understand it, it could be that
of a Paranthropus.
In 1994 Johanson’s team found a very complete maxilla (A.L. 666-1) in
the Hadar region; the allocation of this find to the genus Homo seems
clear. The fossil was found immediately below a volcanic tuff dated at
around 2.3 million years old on the basis of radiometric techniques.
Around twenty stone tools were found together with the human fossil;
this represents the oldest known case of an association of stone tools with
hominid remains. This find has reinforced the hypothesis that it was
representatives of Homo who made the first tools. The oldest known
worked tools were found in another deposit in the Hadar region, Gona,
which is dated at around 2.5 million years old.
Systematic dressing of stone to obtain tools appears to have been one of
the keys to our genus’s ability to inhabit very varied environments and to
gain access to new resources. It is therefore worth leaving bones aside for a
moment to consider stones.

The Stone Cutters

Humans are not the only animals to use tools. Chimpanzees also do so,
but their tools are natural objects suited to their function, such as the
sticks they prepare by stripping away the side shoots and then insert into
the openings of termites’ nests to ‘‘fish’’ for the insects. In these cases the

Hadar Homo

Konso Homo ergaster

Omo Homo habilis

Lake Turkana Homo habilis/rudolfensis

Homo ergaster

Olduvai Homo habilis

Homo ergaster/erectus

Lake Malawi Homo?

Arsuaga / Chosen Species 1405115327-4-007 Final Proof
page 99

Swartkrans Homo ergaster

Sterkfontein Homo habilis?

4.6.2005 2:38pm

Figure 7.1 Location of deposits containing fossils of the first representatives of the genus Homo
Arsuaga / Chosen Species 1405115327-4-007 Final Proof page 100 4.6.2005 2:38pm

A New Kind of Hominid

tool is already formed in nature and no great effort is required to visualize

the final object that needs to be prepared.
Humans are the only primates who really produce tools on the basis of a
shape that exists only in their minds, which they ‘‘impose’’ on the stone. It
is said that the great Renaissance sculptor Michelangelo said of his sculp-
tures, such as David and the Pietà, that they were already enclosed in the
block of marble and all he did was to take away what was not needed. It
was something along these lines, although on a more modest scale, that
humans were doing when they began carving stone.
The first stone tools are known as Oldowan (a name derived from Old-
uvai), and they are very simple. They are classified as Mode 1 tools, and they
consist of pebbles and rocks carved without any standardized form; they
include what are known as choppers (worked on only one side), chopping
tools (worked on two sides), and unworked flakes (Figure 7.2). The process
of manufacturing these tools involves a short sequence of strokes. Un-
modified pebbles and rocks were also used as hammers and anvils.
These simple stone tools would have given their makers something they
lacked, owing to the reduction of their canine teeth: a cutting edge. But

Figure 7.2 Some typical Mode 1 stone tools. Top: two views of a chopping
tool; below: two views of a polyhedron

Arsuaga / Chosen Species 1405115327-4-007 Final Proof page 101 4.6.2005 2:38pm

A New Kind of Hominid

even making such simple tools is more difficult than it seems. Archeologist
Nick Toth and psychologist Sue Savage-Rumbaugh conducted an experi-
ment to teach a male bonobo (pygmy chimpanzee) named Kanzi, who has
shown remarkable capacity for manipulating symbols, to carve stone. (We
shall return to Kanzi later, in the chapter on language.)
In front of Kanzi, Nick Toth struck a pebble with another to obtain
flakes; with the edge of the flakes he then cut a string and opened a box
containing food. Kanzi immediately understood the usefulness of the
flakes and learned to select the best ones and to use them, but when it
came to producing them all his efforts were in vain. There is no doubt that
Kanzi has in his mind the idea of the flake he wants to obtain, but he is
unable to judge how to strike one stone against the other. The reader may
also try to do it: you will see that it is not so easy, particularly if you value
your fingers! In order to carve it, the stone needs to be held steady with
one hand and struck on its edge, slightly at a tangent, with another stone
which acts as a hammer or mallet. Flakes will come away from the
stone which is struck, leaving a carved core. Both the flakes and the core
can be used as tools, because they have edges. Nevertheless, Kanzi
discovered that by throwing a stone against the ground or against a rock
the stone would break apart and produce usable edges, although the result
did not resemble prehistoric tools.
As we have already noted, chimpanzees’ hands are not very well adapted
to manipulating small objects with precision, because of the length of the
palm and all the fingers except for the thumb, the tip of which is some
distance away from the tips of the other fingers. It is a hand made for
hanging from branches, not for carving. However, it is likely that the main
problem lies in the inability to coordinate the necessary movements of the
arms, wrists, and hands.
The australopithecines, at least since Australopithecus afarensis, have
hands which are already very similar to our own, and we may therefore
suppose that they had the mental capacity (just as chimpanzees do) and
the necessary coordination to produce flakes. If they did not do so, it is
perhaps because they did not need sharpened edges to cut with. It may be
that this need arose when the first humans began to eat meat, and
required edges both to cut the thick skin of large animals and to cut
tendons and slice up muscle. They also used stones to break open bones
and extract the marrow, although this activity is similar to that practiced
by chimpanzees when they open nuts using a stone as a hammer and
another, fixed on the ground, as an anvil. The animals that humans ate
would not necessarily have been hunted by them: often they would have
been carrion. They probably also used the sharp edges of stone tools to

Arsuaga / Chosen Species 1405115327-4-007 Final Proof page 102 4.6.2005 2:38pm

A New Kind of Hominid

cut plants, and some believe that what Paranthropus did to prepare tough
plant material using their hyperdeveloped chewing apparatus, the earliest
Homo did with their stone tools.

The Diversification of Homo

The fossil record of Homo from deposits less than 2 million years old is
much richer and more varied. Until a few years ago, all the African fossils
dated between 2 and 1.4 million years old were attributed to two species.
The oldest and most primitive were assigned to the species Homo habilis,
while the more recent and more evolved were seen as African representa-
tives of Homo erectus, a species defined on the basis of fossils found in Java
and China (which we shall consider later). However, since the 1980s
the discovery of fossils with different morphology has led scientists to
rethink the number of species represented by African fossils over one
million years old.
Bernard Wood, who studied the cranial remains and fossil teeth from
Koobi Fora, has been the main agent of this revision. In his opinion, three
species can be distinguished among the oldest human fossils of Africa:
Homo habilis, Homo rudolfensis (the name rudolfensis refers to the old
name for Lake Turkana, Lake Rudolf), and Homo ergaster (ergaster mean-
ing ‘‘worker’’ in Greek). The fossils previously attributed to Homo habilis
are divided between the first two species, while the latter includes virtually
all the remains initially assigned to Homo erectus (with the exception of
skull OH 9, of which more later).
According to Wood, Homo rudolfensis is distinguished by the combin-
ation of a large brain with highly developed masticatory apparatus, similar
in some respects to that of Paranthropus. Homo habilis, on the other hand,
had a brain somewhat larger than that of australopithecines and Paran-
thropus, but smaller than that of Homo rudolfensis, combined with a
primitive overall architecture of the skull very similar to that of Australo-
pithecus africanus. It was distinguished from the latter by a less developed
and shorter masticatory apparatus, the smaller size of the molars and the
presence of a brow ridge separated from the rest of the frontal bone by a
shallow furrow (Figure 7.3).
Nevertheless, in our opinion the main fossils used to justify the exist-
ence of Homo rudolfensis could easily be included in Homo habilis if we
accept that the species showed a marked sexual dimorphism, like the
australopithecines and Paranthropus.

Arsuaga / Chosen Species 1405115327-4-007 Final Proof page 103 4.6.2005 2:38pm

A New Kind of Hominid

Brow ridge

Figure 7.3 Female Homo habilis

Arsuaga / Chosen Species 1405115327-4-007 Final Proof page 104 4.6.2005 2:38pm

A New Kind of Hominid

Let us move on, continuing the assignation of fossils to the species

proposed by Wood in order to complete our account of the information
Homo rudolfensis is a species with a chronological range extending
from 1.9 to 1.6 million years ago, and of which the only fossils known
come from deposits on the shores of Lake Turkana, particularly those
of Koobi Fora. The most significant fossils for this species are skulls
ER 1590, ER 3732, and ER 1470. The most complete of these
is ER 1470, which was found by Richard Leakey’s team in 1972. This
fossil has been dated at 1.9 million years old, and it is the holotype of the
The first fossils attributed to the species Homo habilis were found on
November 2, 1960, during the course of Louis and Mary Leakey’s excav-
ations in Olduvai Gorge. They consisted of part of a mandible, cranial
bones, and bones from the hand of a juvenile. These remains were
numbered as OH 7 and were chosen as the holotype of the species
when Louis Leakey, Phillip Tobias, and John Napier (1917–87) created
it in 1964. Since the first finds, the collection of remains of Homo habilis
from Olduvai has increased substantially, and it now covers the period
between 1.8 and 1.6 million years ago. The skull remains include an
almost complete but very flattened skull known as OH 24 (nicknamed
Twiggy), and an assemblage of skull fragments with mandible numbered
as OH 13 (known as Cinderella or Cindy).
The Olduvai collection also includes a good representation of the
postcranial skeleton, including specimen OH 62, found by Johanson
and White’s team in 1986, which consists of part of the skeleton of an
adult female. Study of the arm and leg bones of OH 62 produced
surprising results. First, estimates of the individual’s height gave a figure
of barely one meter, making it the smallest of all the hominid fossils
(including Lucy). Second, the ratio between the length of humerus and
femur calculated for OH 62 is 95 percent, closer to that of chimpanzees
than that estimated for Lucy herself (85 percent).
Richard Leakey’s excavations at Koobi Fora have yielded, since the early
1970s, the largest and most complete collection of Homo habilis fossils
currently available. The collection includes, in addition to many more
fragmented skull remains, some very complete skulls (ER 1805 and ER
1813), and abundant representation of jawbone fragments and isolated
teeth. Abundant postcranial skeleton fossils have been recovered, includ-
ing an incomplete skeleton (ER 3735) which is similar to OH 62. The
Koobi Fora fossils are concentrated within a short period from approxi-
mately 1.9 to around 1.8 million years ago.

Arsuaga / Chosen Species 1405115327-4-007 Final Proof page 105 4.6.2005 2:38pm

A New Kind of Hominid

In contrast to the rich collection of Homo habilis fossils from Koobi

Fora, the remains of this species found in the Omo River deposits are
limited to isolated teeth and very fragmentary remains which are difficult
to assign to species. One exception is part of a skull (L 894-1) which is
around 1.9 million years old.
To complete this overview, we have one problematic fossil from the
South African Sterkfontein deposit, which is about 1.8 million years old (it
comes from a stratigraphic level more recent than that containing fossils of
Australopithecus africanus). This is a fragmentary skull (Stw 53) which
some researchers assign to Homo habilis, while others see it as belonging
to a species of Homo which has not yet been named.

Ready for the Great Leap

Homo ergaster is distinguished from the preceding Homo species by a

marked increase in brain size, the presence of a brow ridge clearly separ-
ated from the rest of the frontal bone by a pronounced furrow, a position-
ing of the nasal bones such that the nose projects in profile, and a
shortening of the facial skeleton and reduction in the relative size of the
molars (Figure 7.4).
The fossils assigned to Homo ergaster cover the period between 1.8 and
1.4 million years ago. The oldest are from Koobi Fora, and the most
recent could be a mandible from the Konso deposit, discovered in the
same geological level as a set of stone artifacts and the remains of the
Paranthropus boisei we discussed above.
The most notable fossil attributed to Homo ergaster is the skeleton found
by Richard Leakey’s team in 1984 on the bank of the Nariokotome River, a
few kilometers west of Lake Turkana (this is officially numbered WT
15000, but it is familiarly known as Turkana Boy). This skeleton is around
1.5 million years old, and most of the bones of the body are preserved,
including the cranium and the mandible. This discovery is one of the most
important in the history of paleoanthropology, because it is the most
complete skeleton of a hominid of this antiquity. All that is missing are
the bones of the feet and almost all the bones of the hands. On the basis of
the state of development of the bones and teeth, the skeleton has been
identified as a child of between nine and ten years old; the morphology of
the pelvic bones indicates a male. Because the skeleton is so complete, it has
been possible to make a very reliable estimate of his height. The result is
surprising, since it gives a figure of around 1.60 m, suggesting that he

Arsuaga / Chosen Species 1405115327-4-007 Final Proof page 106 4.6.2005 2:38pm

A New Kind of Hominid

Brow ridge

Nasal bones

Figure 7.4 Male Homo ergaster

Arsuaga / Chosen Species 1405115327-4-007 Final Proof page 107 4.6.2005 2:38pm

A New Kind of Hominid

would have been more than 1.80 m tall when he finished growing. Another
very interesting feature which can be reliably established for WT 15000 is
the ratio between the lengths of the humerus and femur (Figure 7.5). This
ratio is 74 percent, entirely human and contrasting with that estimated for
the Homo habilis fossil OH 62 (95 percent).
Two very complete skulls known as ER 3733 and ER 3883 were found
at Koobi Fora. The first of these was found in 1975, in sediments around
1.8 million years old; the second was found a year later in another, more
recent stratum, around 1.6 million years old. Femurs ER 1472 and ER
1481, and coxal bone ER 3228, found at Koobi Fora, and coxal bone OH
28 from Olduvai, may also belong to Homo ergaster.
Outside of East Africa Homo ergaster is represented by an assemblage of
fossils recovered in the South African deposit of Swartkrans (in the same
sediments which contained fossils of Paranthropus robustus, about 1.8
million years old); these include remains of jawbones, teeth, and part of
a skull (SK 847).
Homo ergaster appears as the most human of the early species of Homo.
In addition to its large brain, this species exhibits a height and limb
proportions similar to those of later humans. Moreover, shortly after
Homo ergaster emerged, a new form of stone-dressing appeared, known
as the Acheulian industry.
The Acheulian, or Mode 2 industry, much more elaborate than the
Oldowan or Mode 1 culture, includes cores or large flakes worked on two
sides, known as bifaces and including handaxes, cleavers, and picks. These
tools show a marked degree of standardization in manufacture, and they
require a long sequence of gestures to make them, including turning the
core in the hand while continuing to strike it with the hammer to obtain
flakes. The result is a tool in which all or almost all of its edge forms a
blade. Handaxes are symmetrical bifaces with lateral cutting edges which
converge to a pointed tip; cleavers are bifaces with a straight edge on one
end (Figure 7.6). The handaxes were probably multipurpose tools, used
to cut meat, work wood, and perhaps also to prepare skins. The oldest
Acheulian industry known is around 1.6 million years old, and was found
in the Olduvai deposit in Tanzania.

Family Relationships

One of the most important aspects of a paleontologist’s work is the

attempt to reconstruct the chain of ancestors and descendants who

Arsuaga / Chosen Species 1405115327-4-007 Final Proof page 108 4.6.2005 2:38pm

A New Kind of Hominid

Figure 7.5 Homo ergaster

Arsuaga / Chosen Species 1405115327-4-007 Final Proof page 109 4.6.2005 2:38pm

A New Kind of Hominid

Figure 7.6 Typical Mode 2 tools. Top: two views of a handax; below: two views
of a cleaver

make up the Story of Life. Until we have a time machine to take us back
into the past, no hypothesis of ancestor–descendant relationship can be
confirmed. Nevertheless, it is possible to continue reducing the number of
likely hypotheses, discarding others which are incompatible with three
types of data on the species represented by fossils: their age, their geo-
graphical distribution, and their degree of evolutionary relationship.
It is clear that more recent species cannot be proposed as ancestors of
older species, nor the latter as descendants of the former. Nor can we
assume an ancestor–descendant relationship between species which are
strictly contemporary with one another.
Moreover, in order for a species to be accepted as the ancestor of
another, it is not enough that it be older: it is also essential that it lived
in the same geographical region in which the assumed descendant species

Arsuaga / Chosen Species 1405115327-4-007 Final Proof page 110 4.6.2005 2:38pm

A New Kind of Hominid

Another criterion used by paleontologists when deciding whether to

retain or discard their hypotheses is the degree of evolutionary relation-
ship between the species. The closer the evolutionary relationship, the
more likely it is that there is an ancestor–descendant relationship between
them: the parents of a person can only be found among that individual’s
first-degree relations, and all we need to do is set aside any siblings. If we
knew the degree of evolutionary relationship between all the species of the
past, the problem would be reduced to distinguishing between ‘‘parent’’
and ‘‘sibling’’ species, and this could be done very easily if we also knew
the age and geographical distribution of the species in question.
However, the reality is more complicated, and we have already seen in
previous chapters that it is not always possible to date fossils precisely.
Similarly, we cannot be sure that new fossils will not appear, broadening
the chronological range or area of distribution of a species. Even so, the
thorniest problem for the paleontologist is establishing degrees of evolu-
tionary relationship. At the end of the day, the temporal and geographical
distribution of the species of the past can be established directly on the
basis of fossils; evolutionary relationships, however, are not obvious, but
have to be deduced by the scientists.

The Science of Relationships

Evolutionary relationships are technically known as phylogenetic relation-

ships. In order to deduce these relationships from the fossil record, pal-
eontologists use a method based on very simple principles, implicit in the
Darwinist theory of lines of descent with modification, but much more
complex in application. In essence, the process involves tracking relation-
ship on the basis of similarity between the different species: the greater the
similarity, the closer the evolutionary relationship. But not all resem-
blances are the same. Apart from pure coincidence (such as the fact that
spiders and octopuses have the same number of extremities), similarity
may be due to two main reasons: adaptive convergence (analogy), or
inheritance from a common ancestor (homology).
Adaptive convergence occurs when two anatomical structures resemble
one another not because they were inherited from a common ancestor,
but because they perform similar functions and natural selection has
shaped them in similar ways. One example of this is the fins of sharks
and the flippers of dolphins: they resemble each other because they have a
similar function, but their anatomical structure is very different.

Arsuaga / Chosen Species 1405115327-4-007 Final Proof page 111 4.6.2005 2:38pm

A New Kind of Hominid

Similarities due to inheritance from a common ancestor, on the other

hand, maintain their structural resemblance even though they may show
superficial differences if the organs perform different functions. This may
be seen if we compare the wing of a bat with a dolphin’s flipper, where the
differences are due to the difference in function, but the anatomical
organization is very similar.
In the examples given above, the scientist would reject the superficial
resemblance between the shark’s fin and the dolphin’s flipper, and on the
basis of the structural similarity between the wing of the bat and the
dolphin’s flipper would suggest that the two share a common ancestor
(from whom they inherited the structure of their extremities), and that
this was not the ancestor of the shark; thus, the bat and the dolphin are
more closely related.
This is the first step a paleontologist must take when establishing
relationships between species: distinguishing between common inherit-
ance characters and adaptive convergences. However, common in-
heritance characters are themselves not all the same, nor do they all hold
equal value in reconstructing phylogenetic relationships. We need to
distinguish between primitive characters and derived or evolved characters.
The former do not give us any information on the degree of relationship;
only the latter are useful in establishing whether the relationship between
species is close or more distant. The more derived traits two species share,
the closer their evolutionary relationship.
A good example of this way of working will be found in the chapter
looking at the characteristics of different groups of primates. In that
chapter, we defined the primates with derived traits such as opposable
big toes or the presence of flat nails, both inherited from the exclusive
common ancestor of the primates. At the same time, we eliminate other
characteristics such as the possession of clavicles or five fingers and five
toes, since these are primitive traits of all mammals (and do not indicate
the existence of an exclusive common ancestor for primates). Other
derived traits, such as the absence of a hairless snout, the loss of a
premolar, and adaptations to brachiation, are used to distinguish haplor-
rhines, catarrhines, and hominoids respectively.
Thus the task of the researcher into phylogenetic relationships can be
summed up in two stages. First, distinguishing the convergent characters
from common inheritance characters, and separating primitive from de-
rived traits. Then, the degree of evolutionary relationship between the
species in question can be established on the basis of the presence of
derived traits.

Arsuaga / Chosen Species 1405115327-4-007 Final Proof page 112 4.6.2005 2:38pm

A New Kind of Hominid

This is not always an easy task, given the complexity of the evolutionary
process and above all the scarcity and state of preservation of the fossils
themselves. These difficulties mean that different researchers arrive at
different conclusions on the nature of the characters used in the analysis.
As a result, there is no consensus on the hominid evolutionary tree, and
various hypotheses have been put forward by different authors.

The Hominid Tree

If we were to describe all the evolutionary trees put forward for hominids,
explaining how the appearance of new fossils has progressively eliminated
some and resulted in the presentation of other new theories, we would no
doubt try the patience of our readers. We therefore prefer to discuss the
evolutionary tree which in our view is the most compatible with the
evidence currently available (Figure 7.7).
In view of its antiquity and the primitive nature of the majority of its
traits, the species Ardipithecus ramidus appears the ideal candidate for the


P. boisei P. robustus
H.habilis/ H. ergaster
H. rudolfensis
A. africanus
Years (millions)

First Homo
P. aethiopicus

−3 ?

A. afarensis

A. anamensis

A. ramidus


Figure 7.7 The authors’ evolutionary diagram for the first hominids

Arsuaga / Chosen Species 1405115327-4-007 Final Proof page 113 4.6.2005 2:38pm

A New Kind of Hominid

role of ancestor of the other hominids. Until we have information on this

species’ mode of locomotion, the only derived hominid characteristic
found in Ardipithecus ramidus appears to be the morphology of the
upper canine and the third lower premolar.
More recent than Ardipithecus ramidus, Australopithecus anamensis
shows at least one other derived trait: the thick layer of enamel on the
molars. (If Ardipithecus ramidus is eventually shown not to have been
bipedal, bipedalism would be another new feature in Australopithecus
anamensis.) We do not yet know enough about the anatomy of Australo-
pithecus bahrelghazali to be able to include it with confidence in the
hominid evolutionary tree.
The majority of authors consider Australopithecus afarensis to be the last
common ancestor of later hominids (Figure 7.7). However, we take a
different view. Our studies on the region at the base of the skull related to
mastication have led us to conclude that the chewing apparatus of Aus-
tralopithecus afarensis was proportionately wider than that of apes, austra-
lopithecines, and humans, but equivalent to that of Paranthropus. The
widening of the masticatory apparatus, which is reflected in the presence
of relatively wide jawbones, is characteristic of Paranthropus and tends to
favor circular movements of the mandible, necessary for processing tough
foodstuffs. In our opinion, this resemblance between Australopithecus
afarensis and Paranthropus is not a phenomenon of adaptive convergence
but a shared derived character. Consequently, we believe that Australopith-
ecus afarensis already formed part of the lineage of Paranthropus and could
not have been a direct ancestor of the Homo line (Figure 7.7).
The Paranthropus branch is characterized by a number of derived traits,
mainly related to the particular specialization of its masticatory apparatus.
Nevertheless, there is a gradation within this specialization which helps us
to establish the relationship between the three known species of Paran-
Thus, in aspects such as its high degree of prognathism (compare
Figure 6.3 and Figure 6.4), and a number of details of the morphology
of the base of the skull, Paranthropus aethiopicus appears to be the least
specialized of the species. This more primitive cranial anatomy, combined
with its greater age, makes this species the ideal candidate to be considered
the ancestor of the other Paranthropus species.
The first representatives of the genus Homo are very similar to the
specimens of Australopithecus africanus, from which they are distin-
guished by their larger brain, somewhat less developed masticatory appar-
atus and the presence of a slight brow ridge. Apart from these differences,
the architecture and morphology of the rest of the cranium are very

Arsuaga / Chosen Species 1405115327-4-007 Final Proof page 114 4.6.2005 2:38pm

A New Kind of Hominid

P. boisei P. robustus
H.habilis/ H. ergaster
H. rudolfensis
A. africanus
Years (millions)

First Homo
P. aethiopicus

−3 ?

A. afarensis

A. anamensis

A. ramidus


Figure 7.8 Alternative evolutionary diagram for the first hominids

similar in the two species (compare Figure 4.4 and Figure 7.4). This seems
to us to indicate a very close relationship between Australopithecus afri-
canus and Homo. In our view, the common ancestor of the two lived
between 4 and 3 million years ago.
As far as evolutionary relationships between species in our own genus
are concerned, the question of whether Homo rudolfensis existed or not
makes very little difference. If we recognize only the existence of Homo
habilis the problem is somewhat simpler, since only two species remain to
be related, and on the basis of both chronology and cranial anatomy Homo
habilis could be the ancestor of Homo ergaster.
In any case, if we agree the existence of Homo rudolfensis, the species has
to be considered an evolutionary offshoot within the Homo genus, since
the anatomy of its masticatory apparatus is too specialized to propose it as
the ancestor of the other species of the genus. The position of Homo
habilis as the original species of Homo would thus remain unchallenged.
Our evolutionary tree for hominids implies that Paranthropus origin-
ated in East Africa. From there, representatives of this lineage reached the
ecosystems of the south of the continent, where they evolved separately
from their East African fellows and gave rise to a different species,

Arsuaga / Chosen Species 1405115327-4-007 Final Proof page 115 4.6.2005 2:38pm

A New Kind of Hominid

Paranthropus robustus. This phenomenon of local evolution of a periph-

eral population has been common in the history of hominids, and we will
encounter it again later on in the evolution of Homo.
Since the oldest fossils of our genus come from the east of the contin-
ent, the simplest hypothesis is that it was there too that our genus
originated, and from there spread to the more southerly regions of Africa.
Now that we have described the fossil record of the oldest hominids,
examined their basic adaptations and discussed their family relationships,
it is time to look at other questions of their life, such as the size of their
brains, their diet, their development, sexuality, and sociability.

Arsuaga / Chosen Species 1405115327-4-008 Final Proof page 116 4.6.2005 2:39pm

The Evolution of the Brain

I thence concluded that I was a substance whose whole essence or nature

consists only in thinking.
René Descartes, Discourse on Method

The Organ of Intelligence

Human beings are characterized by a much more developed intelligence

than other animals. Although what we call ‘‘intelligence’’ is a concept that
is difficult to define and very problematic in terms of measurement, it is
clear that it is related to certain skills which are unique to us. No other
animal has yet proved capable of writing a book, composing a symphony,
traveling to the moon, designing and constructing a computer, or simply
wondering about its origin and its destiny.
The region of our organism devoted to performing these functions is
the brain, which is housed in the interior of the cranial cavity and is
composed of three organs, the cerebrum, the cerebellum, and the brain-
stem (Figure 8.1). In very general terms, we can say that the superior
functions related to intelligence (for example the capacity for abstraction,
the association of information, or the ability to encode and decode
information via an articulated language) are performed in the cerebrum,
which is divided into two hemispheres, left and right. The cerebellum is
responsible for general motor coordination and balance, while the brain-
stem acts as our ‘‘automatic pilot,’’ responsible for maintaining regular
functions such as breathing and heartbeat.

Arsuaga / Chosen Species 1405115327-4-008 Final Proof page 117 4.6.2005 2:39pm

The Evolution of the Brain


Wernicke's area

Broca's area



Figure 8.1 The human brain The areas of the cerebral cortex directly related
to language are shaded

Research into the evolution of the brain in hominids has focused on

two main areas: evaluating the increase in the size of the brain itself
over the course of hominid evolution (the process of encephalization),
and analyzing the morphology of the organs of the brain, especially
the cerebrum.

Large and Small Brains

When, in our lectures, we ask which is the animal species with the largest
brain, almost all students immediately reply: humans. The conviction that,
in line with our superior intelligence, our species is the most encephalized
in the animal kingdom, and that this is the basic characteristic that
distinguishes us from other animals, is embedded in our culture. How-
ever, this is not the case, or at least not in absolute terms.
The average size (mass) of the human brain is around 1,250 g,
and although this is greater than that of any other primate species

Arsuaga / Chosen Species 1405115327-4-008 Final Proof page 118 4.6.2005 2:39pm

The Evolution of the Brain

(chimpanzees and gorillas have average brain mass of around 400 and
500 g, respectively), and of the majority of animals, it is markedly less than
that of the large mammals, whose brains are much larger than ours: the
blue whale (the largest animal that has ever existed) has a brain mass of
around 6,800 g, while the African elephant (the largest living land mam-
mal) has a brain mass of about 5,700 g.
Does this mean that blue whales and elephants are more encephalized,
and therefore more intelligent than humans? Of course not. What hap-
pens is that since the work of the brain is to coordinate the functions of
the rest of the body, larger bodies predictably require larger brains to
ensure sufficient coordination. Thus when we compare the brain size of
two species we have to consider the influence of body size on the size
of the brain.
A simple way of relating body size to brain size is to divide brain mass by
body mass; in this way we can find the ratio between the two. The greater
the ratio, the more brain an animal would have per unit of body mass. On
first view, this might seem the most appropriate way of comparing the
brain size of animals of different sizes. However, this method also fails to
reveal humans as the most encephalized species in the animal kingdom.
Paradoxically, we are overtaken by the smallest mammals, which have
proportionately larger brains.
Should we then resign ourselves to accepting that the smallest mammals
are more encephalized than we are? We can rest easy: shrews do not have a
more developed brain than we do. The truth is that brain size increases
more slowly than body size: in other words, the size of the brain becomes
proportionately smaller in larger mammals. This phenomenon, the
change in the proportions of the organs as body size increases, is very
common in living beings, and was discovered in the 1920s; it was named
allometry, or the Law of Disharmony.
Allometry may occur during the growth of an organism: thus, for
example, a newborn has a proportionately bigger head than an adult,
and its limbs are relatively shorter. If a baby maintained these proportions
as it grew, the result would be an adult of very alarming appearance
(Figure 8.2).
Allometry may also occur between adults of the same species. A good
example of this is the variation in ‘‘ideal weight’’ in relation to height. As
height increases, the ideal weight becomes relatively greater: for a woman
150 cm tall the ‘‘ideal weight’’ is 47 kg (315 g per centimeter), while
another woman 180 cm tall has an ‘‘ideal weight’’ of 67 kg (372 g per
centimeter). But the kind of allometry which most interests us is the variation
between species in the same group. Thus in mammals, brain size varies in

Arsuaga / Chosen Species 1405115327-4-008 Final Proof page 119 4.6.2005 2:39pm

The Evolution of the Brain

Figure 8.2 Change in proportions of body parts (allometry) during growth.

Left : a newborn; center: a three-year-old child; right : an adult – all drawn the same

relation to body size, and as we have already noted, brain size becomes
proportionately smaller as body weight increases. While in chimpanzees
the ratio between brain mass and body mass averages 0.01 (one gram of
brain per 100 g of body weight), in the mouse lemur (a small primate from
Madagascar with a body mass of only 60 g and brain mass of 1.8 g) the ratio
is around 0.03 (i.e. 3 g of brain for each 100 g of body weight).
Given that mammals’ brains grow allometrically in relation to body size,
the only way of comparing the brains of different species is to calculate the
brain mass that each should have in relation to its body mass, and to compare
these ‘‘ideal brain masses’’ with their actual brain mass. The species with the
greatest surplus brain mass would be the most encephalized.
Clearly, the key to comparing species of different size lies in calculating
the expected values of the brain mass of each. To this end the mathemat-
ical law which relates the body mass of an organism to its ‘‘ideal brain
mass’’ has been calculated, on the basis of analysis of extensive data on
brain mass and body mass in a large number of mammal species.

Arsuaga / Chosen Species 1405115327-4-008 Final Proof page 120 4.6.2005 2:39pm

The Evolution of the Brain

World Champions of Encephalization

The brain mass which would correspond to an organism’s body mass is

the expected value (its ‘‘ideal brain mass’’). We can call the real size of its
brain the actual value. The ratio of the two values (actual/expected) is
known as the index of encephalization, and it measures the disparity
between the size of brain an animal should have and the actual size of its
When the index of encephalization for a species is equal to 1, we can say
that its expected and actual values are equal and it therefore has the brain
which corresponds to its body size. If the index is less than 1, the actual
value is less than that expected and the species has a smaller brain than it
should have according to its body mass, while an index higher than 1
indicates a brain larger than expected.
Various authors have calculated indices of encephalization for a large
number of mammal species, and their results agree on a number of points.
First, primates appear as a highly encephalized group of mammals (but
not the most encephalized, contrary to popular belief). In virtually all the
primates which have been studied, the value calculated for the index of
encephalization is greater than 1.
The second striking result is the fact that Homo sapiens is the most
encephalized of all mammals, with an index of more than 7. In other
words, our brain is more than seven times the size that could be expected
in a mammal of our body mass.
Finally, one further singular fact: the species shown to be closest to
humans in terms of encephalization are not primates but cetaceans
(whales and dolphins), in particular dolphins, which have indices of ence-
phalization above 4.
These results bear out the generalized belief that humans are the most
encephalized species of the animal kingdom. But is this a characteristic
common to all hominids?

Weighing Ghosts

Although the brain does not become fossilized, the skull, which contains
the cavity which houses it, does. In skulls where the whole or a good part
of the neurocranium is preserved, it is possible to measure the volume of
the cavity which houses the brain. Since the brain occupies virtually the

Arsuaga / Chosen Species 1405115327-4-008 Final Proof page 121 4.6.2005 2:39pm

The Evolution of the Brain

whole of this space, the volume of the cranial cavity (cranial capacity) can
be used as a measure of the volume of the brain itself. Experimental
studies on primates and humans have shown that cranial capacity is almost
directly equivalent to brain mass: a capacity of 1,000 cc corresponds to a
mass of 971 g.
Estimating body mass from fossils is a different matter. Here the prob-
lem is more complicated because there is no measurement we can take
from the skeleton which would directly correspond to the individual’s
body mass. In tackling this problem, paleontologists start from an obvious
fact: the size of the bones is related to the size of the body itself, so large
bones should correspond to similarly large individuals.
However, not all the bones of the body follow this rule. For example,
we have seen that Paranthropus had a highly developed jawbone,
owing to its particular adaptation to diet. So if we applied the rule
that large bones (and teeth) reflect large body size to the jawbones of
Paranthropus, we would have to conclude that these hominids had a
body size close to that of a female gorilla. This mistake was in fact made
in the past, and is the source of the adjective ‘‘robust’’ usually given to
In order to arrive at reliable estimates of the body mass of fossil
hominids, we need to use bones or parts of bones which are clearly
involved in supporting body weight. Modern humans have several
bones which satisfy this requirement. In our species the weight of the
trunk, the upper limbs and the head is transmitted through the lumbar
region of the spine to the sacrum, in the pelvis. From here it passes
alternately at each stride to each of the femurs (which also have to support
the weight of the leg which is in the air), and through them down to the
bones of the feet (which support virtually the entire body weight at each
step). The lumbar and sacral vertebrae, the coxal bones and the bones of
the lower limbs are thus good candidates for estimating an individual’s
body mass.
Nor are all the dimensions of a bone equally suitable for estimating
body mass. For example, the length of the femur has more to do with an
individual’s height than his body mass. Since the body weight is transmit-
ted from one bone to another through the joint surfaces of the two bones,
these are directly related to body mass. The pressure exerted on a joint is
directly proportional to body mass and inversely proportional to the
joint’s surface area; therefore, in order to avoid the pressure on a joint
increasing with an increase in weight (and consequent injury to cartilage
and bone tissue), a correlative increase in surface area is required. Thus,
large joint surfaces indicate high body mass.

Arsuaga / Chosen Species 1405115327-4-008 Final Proof page 122 4.6.2005 2:39pm

The Evolution of the Brain

Having identified the bones and the anatomical regions which are
suitable for calculating body mass, there is still another problem to re-
solve. What is the mathematical relationship between the joint surfaces of
each bone and the body mass?
This question can be approached by analyzing data on joint surfaces and
body mass in living primates, in order to determine the mathematical
function relating them. But we have already seen that not all primates
walk in the same way, nor do they have the same body proportions, and
these factors affect how the body weight is transmitted. For example, in
apes, which support themselves on four limbs when they move over the
ground, the humerus is more involved in the transmission of weight,
while the femur supports less weight than in modern humans. Given
that the first hominids were fully bipedal, comparison with our own
species seems most appropriate. Nevertheless, as we have noted in previ-
ous chapters, the body proportions of australopithecines and Paranthro-
pus appear to have been intermediate between those of apes and our own.
Thus the data on these primates also needs to be taken into account when
estimating the body mass of fossil hominids.
Paleontologists meet a further obstacle when estimating the weight of
bygone hominids: the scarcity of the fossil record itself. Few even roughly
complete skeletons of the first hominids are known, and only in one or
two cases has it proved possible to deduce an individual’s mass on the
basis of a range of bones (allowing us to refine the calculation). We also
have very few bones suitable for use for each species, and therefore a
very limited number of estimates for any one species. One final added
difficulty is the fact that it is not always easy to assign a bone of the
postcranial skeleton, found in an isolated position, to a specific species
of hominid.
Despite all these problems, researchers such as Henry McHenry and
William Jungers have undertaken a series of investigations which, using
the available evidence and a variety of statistical techniques (together with
a good dose of common sense), have yielded comparable results.
Australopithecines and Paranthropus appear to have had very similar
body mass. Let us recall that the average body mass of male and female
Australopithecus afarensis, the species for which body mass has been most
accurately estimated, was 45 kg and 30 kg, respectively. Homo habilis does
not appear to have been any bigger than the australopithecines and
Paranthropus, whereas the values estimated for Homo ergaster are over
55 kg, indicating a marked increase in body mass toward values similar to
those of our species.

Arsuaga / Chosen Species 1405115327-4-008 Final Proof page 123 4.6.2005 2:39pm

The Evolution of the Brain

The Brain Size of Fossil Hominids

Having estimated body mass, all we need to know, in order to calculate

the indices of encephalization of fossil hominids, is their brain mass. The
average brain mass1 in the various species of hominid is as follows: Aus-
tralopithecus afarensis, 426 g; Australopithecus africanus, 436 g; Paran-
thropus robustus, 523 g; Paranthropus boisei, 508 g; Homo habilis/Homo
rudolfensis, 619 g; Homo ergaster, 805 g.
Using all of this, we can calculate the indices of encephalization of the
first hominids. These are as follows: Australopithecus afarensis, 1.3; Aus-
tralopithecus africanus, 1.4; Paranthropus boisei, 1.5; Homo habilis/rudol-
fensis, 1.8; Homo ergaster, 1.9. The sample for Paranthropus robustus is
too small to make a satisfactorily reliable calculation of the index of
On the basis of these values, we can draw some very interesting con-
clusions. First, australopithecines and Paranthropus show an index of
encephalization higher than that of chimpanzees (1.2), but markedly
lower than that of the first representatives of Homo, which have indices
of encephalization 50 percent higher than that of chimpanzees and almost
two-thirds that of modern humans (2.9).2 Second, the increase in brain
size noted in Homo ergaster compared to Homo habilis was compensated
by a proportionate increase in body size, resulting in very similar indices of
encephalization for the two species.

Although we can calculate an average brain mass for each of the different species,
the ranges of variation may overlap, with the higher values in one species overlapping
with the lower values for another. This is the case, for example, with Paranthropus boisei
and Homo habilis: the former has an average brain mass of 508 g, but a range of
variation from 470 g (fossil ER 13750) to 590 g (Konso specimen). Homo habilis,
on the other hand, has an average of 619 g but includes specimens with a brain mass of
only 503 g (such as ER 1813) and others up to 661 g (OH 7), or 736 g, if fossil ER
1470 is included in this species (as it is for the purpose of our calculations).
Our calculation of indices of encephalization was based on a different formula from
that used in the section where we compared the encephalization of all mammals, which
gave a value of 7 for our species. For this calculation we used the formula which
estimates ideal brain mass on the basis of data from haplorrhine primates. This new
calculation gives a value of 2.9 for our species. This means that we have a brain mass
seven times the expected value for a mammal of our body mass, but only 2.9 times
greater than that which would correspond to a haplorrhine primate of our size.
Remember that primates are highly encephalized mammals.

Arsuaga / Chosen Species 1405115327-4-008 Final Proof page 124 4.6.2005 2:39pm

The Evolution of the Brain

Thus a marked increase in encephalization occurs around the origin of

the genus Homo. But can this quantitative leap alone account for the
greater intellectual complexity of the first humans, especially from Homo
ergaster onward? Were the brains of the first humans simply bigger than
those of australopithecines and Paranthropus, or were they also different
in their organization? In order to answer these questions, we need to take
a look at the part of the brain responsible for what we call intelligence: the

Surface Area of the Brain

The brain is made up of two different substances: white matter, which

forms the internal part of the brain and the majority of its volume, and
gray matter, which is almost entirely confined to a thin superficial layer, or
cortex. Gray matter is made up of the bodies of the neurons (the highly
specialized cells which form the nerve tissue), while white matter consists
of the extension of these neurons (axons) which serve to connect and
relate the neurons.
The cerebral cortex of non-mammalian vertebrates is very limited in
area, and deals only with the analysis of olfactory stimulation; it is known
as the paleocortex. In mammals the cerebrum is more highly developed
and a new zone, the neocortex, appears on top of the paleocortex. In apes
and humans the neocortex makes up virtually all of the cerebral cortex.
The surface of our brain is not smooth; it has a very complex topog-
raphy in which a series of furrows (in fact folds in the cerebral cortex),
known as sulci, can be seen; these run around mounds known as gyri.
When the sulci are very marked they are known as fissures; these separate
broad zones of the brain, or lobes. There are four lobes in each cerebral
hemisphere: frontal, temporal, parietal, and occipital. The lobes are lo-
cated in roughly the same region as the bones which bear the same names
(Figure 8.3).
The brains of humans and apes are specialized in function, and we can
locate regions associated with specific tasks in the surface of the brain.
These functional maps can be related to the surface topography of sulci,
gyri, fissures, and lobes, so that by examining the brain surface we can
analyze the relative degree of development of each specific functional area
and attempt to relate this to the skills and behavior of the animal.
Although there are numerous morphological differences between the
human brain and that of apes, owing to their different kinds of mental

Arsuaga / Chosen Species 1405115327-4-008 Final Proof page 125 4.6.2005 2:39pm

The Evolution of the Brain

Frontal lobe

Parietal lobe

Occipital lobe

Temporal lobe

Figure 8.3 Sketch of a human brain, showing the lobes

activity, we can only track the evolution of a few of these in fossils because
our knowledge of the brain surface of fossil hominids is limited by various
First, we have no fossil brain, only casts (natural or artificial) of the
cranial cavity, and although the internal walls of the cranium reproduce
the general morphology of the brain surface they do not do so with
sufficient detail to identify the exact limits of most areas of the brain
with any precision. This is because the brain is covered by a triple layer
of membrane, the meninges.
Moreover, the larger the brain of an organism, the more faint and
indistinct the impressions of the brain on the inner surface of the cranium.
Some regions of the brain surface also leave more detailed impressions
than others. Thus, for example, the impressions of the occipital lobe
are lighter than those of the frontal lobe. In short, studies of the
paleoneurology of the first hominids are limited to a few specific aspects
of brain morphology.
The two world authorities in this area, Ralph Holloway and Dean Falk,
hold different opinions on the brain of australopithecines and Paranthro-
pus. Holloway believes that these hominids already show evidence of
reorganization of the brain compared with apes; these are reflected, for
example, in a degree of development of Broca’s area (an area related to the
production of speech, which we shall look at in the chapter dedicated to

Arsuaga / Chosen Species 1405115327-4-008 Final Proof page 126 4.6.2005 2:39pm

The Evolution of the Brain

language). However, Falk takes the view that the brain surface of Paran-
thropus and australopithecines is clearly similar to that of apes and that
there is no evidence of reorganization of the brain.
Despite their differences on this matter, Holloway and Falk agree that
there are clear indications of cerebral restructuring, similar to that of
modern humans, in the fossils of the first representatives of Homo.
These indications can be summed up in two major processes: the appear-
ance of marked asymmetry between the cerebral hemispheres, and a more
complex morphology of the frontal lobe.
The human brain is clearly asymmetrical, and although apes, and an-
thropoids in general, show cerebral asymmetries, these are not of the same
kind or as extensive as those typical in humans. The asymmetries charac-
teristic of the human brain are not directly related to the functional
specialization of the cerebral hemispheres observed in humans. This phe-
nomenon is known as brain lateralization, and seems to be confined to
our species. To generalize, the left hemisphere specializes in functions
related to language, precision movements of the right hand, the capacity
for analysis, and the perception of temporal sequence. The right hemi-
sphere is devoted more to tasks such as the skill of the left hand, the
capacity for overall understanding of processes, recognition of faces,
spatial vision, musical skills, the control of the tone of the voice, and the
expression of emotions and recognition of them in others.
In addition to other functions (such as control of primary motor
functions), the frontal lobe is responsible for a series of mental capacities
which are exclusive to or most fully developed in humans. These include
the capacity to determine the sequence of movements of the phonetic
apparatus which make up speech, control of the emotions, the ability to
concentrate on a task, planning, anticipation of events, maintaining an
idea in the mind for a long time, and control of the use of the memory to
integrate previous experiences and learning when taking decisions. Meta-
phorically speaking, we could say that the frontal lobe is the ‘‘conductor’’
of the orchestra of our brain.
Over the course of human evolution the frontal lobe has expanded a
great deal, both in absolute terms and in relation to the rest of the brain.
But recent investigations by a group of scientists led by Katerina Semen-
deferi indicate that the volume of our frontal lobe is what would be
expected in a primate with a brain the size of ours. Given that the frontal
lobe appears to be the most ‘‘human’’ region of our brain, perhaps the
increase in brain size is merely the result of the selective advantage that
having an increasingly more developed frontal lobe conferred on our
ancestors. In other words, it is reasonable to suppose that the large size

Arsuaga / Chosen Species 1405115327-4-008 Final Proof page 127 4.6.2005 2:39pm

The Evolution of the Brain

of the brain in our species is not the cause, but the result of our large
frontal lobes. If this is the case, development of the frontal lobe would
have been the driving force in our encephalization.
What we can at least say with certainty is that the surface of the frontal
lobe has become increasingly complicated, owing to an increase in the
number of sulci which cannot be explained purely in terms of the greater
size of the human brain.
These characteristics – the cerebral asymmetry linked to functional later-
alization, and greater structural complexity of the frontal lobe – are clearly
marked in the endocranial casts of Homo habilis/Homo rudolfensis and
Homo ergaster. So even in the first humans, in addition to an increase in
brain size, we find clear indications of a brain structure similar to our own.
To complete our examination of the brain of the first hominids, we
might speculate as to the nature of the vital advantage the new type of
brain gave our ancestors, favoring their selection. There are two hypoth-
eses on this subject.
The first, put forward by Robin Dunbar and Leslie Aiello, links the
increase in the neocortex (i.e., the cerebrum) to the improvement in social
skills within the group. According to these authors, the increase in size
and the reorganization of the human brain are linked to the development
of ‘‘social intelligence.’’ We shall return to this hypothesis when we look at
the social biology of the first hominids.
Falk, on the other hand, relates the modifications in the brain of the first
humans to a very specific capacity, that of language. She argues that given
that one of the crucial centers of human speech is located in the frontal lobe,
and that the production and decoding of language are functions which are
clearly lateralized in the human brain, both lateralization and the increased
structural complexity of the frontal lobe are related to the development of
the linguistic capacities of our first ancestors. It is immediately evident that
these two hypotheses are complementary, since the primary social skill is
perhaps the capacity to communicate efficiently with others.
However, the linguistic capacities of the first humans are the subject of
major controversy. We shall return to this subject in the chapter on

The Size of the Intellect

During the last quarter of the 19th century it was fashionable in some
scientific circles to link human intelligence with brain size. During that

Arsuaga / Chosen Species 1405115327-4-008 Final Proof page 128 4.6.2005 2:39pm

The Evolution of the Brain

time, a series of measurements of brain size was made on a number of

skulls and cadavers, with the aim of identifying the variations in intelli-
gence in humans. These studies were principally the work of Paul Broca
(1824–80), Professor of Surgery at the Faculty of Medicine in Paris, and
his followers. One of the most significant results of these studies was the
finding that on average, women have smaller brains than men. This fact
was used to assert that, because of their smaller brain mass, women are
intellectually inferior to men.
Today, at the end of the second millennium, this story might appear
outdated to many of our readers. However, the argument that women’s
smaller brain mass indicates reduced intelligence still persists to some
extent today.
As we have seen in this chapter, it is not overall brain size which
determines the size of the intellect, but the proportion of brain size to
body size, or the degree of encephalization. Women’s brains are, on
average, smaller than those of men because women’s bodies are also,
on average, smaller. Both women and men have brains which correspond
to their respective body size; in other words, the two sexes are equally

Arsuaga / Chosen Species 1405115327-4-009 Final Proof page 129 4.6.2005 2:39pm

Teeth, Guts, Hands, Brain

Compare, at entirely similar ages, a man who, in order to free himself

for studies and habitual intellectual work, has acquired the habit of
eating very little with another who habitually takes plenty of exercise,
frequently goes out of the house, and eats well. The stomach of the first
will have very little capacity and will be filled by a very small quantity
of nourishment, while the stomach of the second will have preserved and
even increased its capacity.
Jean Baptiste de Lamarck, Zoological Philosophy

Types of Diet

There is ongoing debate in our society as to whether we are ‘‘naturally’’

vegetarians or carnivores. In fact many campaigning vegetarians include
milk and dairy products, which are derived from animals, as well as eggs,
in their diet. Given that, on the other side, there is virtually no one who is
exclusively carnivorous by conviction, the debate is generally between
ovo-lacto-vegetarians and omnivores, who in addition to eating veget-
ables, eggs, and dairy products also eat meat and fish. Before going
further, we must make it clear that both diets have been scientifically
proven to be healthy, provided that they are balanced. However, most
specialists would not say the same of a strict vegan diet.
Nevertheless, there are those who take the view that as apes, our diet
should consist only of vegetables, ignoring the fact that many primates eat
greater or lesser quantities of invertebrates, particularly insects, and small

Arsuaga / Chosen Species 1405115327-4-009 Final Proof page 130 4.6.2005 2:39pm

Teeth, Guts, Hands, Brain

vertebrates. But since primates vary so widely, perhaps we should restrict

ourselves to looking at our closest relatives. It is true that the diet of the
great apes consists almost exclusively of vegetables, mainly ripe fruit,
leaves, stems, and tender shoots. Orangutans and chimpanzees eat more
fruit, or are more frugivorous, than gorillas, which eat more leaves,
shoots, and stems, and are thus more folivorous.
Everything has its advantages and disadvantages; fruits are considerably
more nutritious because they contain more simple carbohydrates like
glucose, fructose, and sucrose. Leaves and tender stems are probably
more abundantly available, but contain more fiber, which is carbohydrate
that we cannot assimilate, such as cellulose (although fiber is also import-
ant for good digestion). However, although no one would classify chim-
panzees as predators, they have been seen hunting monkeys and young
antelope and other small mammals. They also eat insects such as caterpil-
lars, ants, and termites. To sum up, while they may be basically vegetarian,
they seem to like animal proteins.
In the case of our species we cannot rely on ‘‘field’’ studies to resolve
the problem of man’s natural diet in his environment. Since agriculture
and animal husbandry were ‘‘invented’’ during the Neolithic period,
10,000 years ago, animal and plant foods have been eaten in varying
proportions depending on the culture in question. Nevertheless, the
majority of humanity has lived, and continues to live, principally on the
hard grains of cereals – wheat, rice, corn, barley, oats, rye, millet, and so
on. Another important element in human sustenance is the dry seeds of
legumes (beans, peas, lentils, etc.). All of these grains and seeds (cereals
and legumes) are rich in long-chain carbohydrates (particularly starch),
which are made up of many molecules. Legumes contain more protein
(around 20 percent of their mass) than cereals (about 10 percent), but in
both cases these are low-quality proteins in the sense that they are incom-
plete. Let us look at what this means.
Proteins are long chains of molecules known as amino acids. Of the 20
types of amino acids which exist in proteins, humans can only synthesize 11.
We have to obtain the others, known as the essential amino acids, from food.
Five of these amino acids are less abundant in foodstuffs, and some are not
found at all in some cereals and legumes, although all can be obtained from
an appropriate combination of various types of these plant products.
It is worth pointing out that cereals and legumes are not eaten as they
are found in nature: they need to be prepared. The minimum required to
prepare them is tools for grinding, to produce flour, and fire, water, and a
ceramic vessel to cook and soften them. Thus to some extent they are not
‘‘natural’’ foods.

Arsuaga / Chosen Species 1405115327-4-009 Final Proof page 131 4.6.2005 2:39pm

Teeth, Guts, Hands, Brain

The remainder of the vegetable element in the human diet consists of

tubers, like potatoes, which are rich in starch, and vegetables. Vegetables
comprise different parts of plants, from leaves (like lettuce and spinach),
to bulbs (such as onion, garlic, and leeks), roots (carrots or beet), flowers
(cauliflower and artichokes), stems (like celery), and fruit (such as toma-
toes and pimentos). Vegetables contain a lot of water and few calories.
Finally, we have fruit. Although some vegetables are eaten with little
preparation, only fruit is eaten as found in nature, without adding any-
thing or cooking it.
Nuts, which are fruits with shells (such as almonds, hazelnuts, and
walnuts), and which are very nutritious because of their high fat content,
but fairly well protected against our fragile teeth, merit separate mention.
Incidentally, while a hazelnut is a nut in the botanical sense, almonds and
walnuts are in fact fruits with an internal stone which contains the nutri-
ents (technically they are drupes, or stone fruit, like cherries and olives,
except that in the case of the latter we eat the fleshy envelope).
The total amount of protein we require is fairly small: even highly
trained athletes need no more than 1.5 g per kilogram of body weight,
and for most of us the total required is less than 100 g. Thus we do not
need to eat a large quantity of animal products, and the amount consumed
in First World countries is clearly excessive.
So we see that there are major differences between the dietary habits of
apes and our own, at least since Neolithic times. Comparing apes’ diet
with human diet, we could say that apes eat vegetables and fruit, while we
(as a species) live on the tough grains of cereals – a form of grass – and the
seeds, also dry, of legumes.
There are some human groups that have been identified and studied
within recorded history which do not engage in agriculture or animal
husbandry. It is worth noting that they hunt, fish, and gather very varied
animals and plants. What they eat varies in relation to a number of factors,
such as the resources available in the environment and the time of year,
but no cases of groups living either exclusively on what they hunt or
exclusively on plant foods have been found. The only exception is perhaps
the Inuit, whose economy until recently relied almost entirely on hunting
and fishing; however, this is an exceptional case related to the special
characteristics of their environment, which is covered in ice and snow
for much of the year and devoid of any vegetation.
How do we approach this question from the paleontological point of
view? By studying the only part of the body directly related to diet which
fossilizes – the teeth, and together with them, the chewing apparatus.

Arsuaga / Chosen Species 1405115327-4-009 Final Proof page 132 4.6.2005 2:39pm

Teeth, Guts, Hands, Brain

Carnivorous and Herbivorous Mammals

Carnivorous mammals have adapted their teeth to the functions they need
them to perform. Carnivores vary widely in form, but all of them have
certain teeth specialized in cutting meat, known as carnassial teeth (these
are the last upper premolar and the first lower molar). Carnassial teeth
have a very sharp edge which allows them literally to slice up the flesh of
dead animals (Figure 9.1).
Proteins and animal fats are rich in energy, having a high calorific value,
and they are also easy to digest. This means that this type of food does not
need to be prepared in the mouth before it is swallowed, so chewing is
brief, its purpose being only to reduce the flesh of the dead animal to
pieces of a size that can pass through the esophagus. The carnassial teeth
operate like shears, or scissors, the upper one passing to the outside of the
lower when the mouth is closed. This scissor-like movement causes the
carnassial teeth to sharpen themselves on one another, and as they wear
against one another they become razor-sharp. Unlike our kitchen knives,
this wear improves the cutting edge of the carnassial teeth and increases
their efficiency, rather than dulling or chipping them.
Carnivores also need tools to kill their prey, and the canines, often
known as eye-teeth, are highly developed in order to perform this func-
tion. The terrestrial carnivores with the most highly developed canines are
the Felidae, the family which includes cats, lions, tigers, and so on. In
these animals the incisors are small and the canines very large; the carnas-
sial teeth are highly developed and are the hindmost teeth in the jaw
(although the vestige of a molar remains behind them).
Hyenas have a similar dentition to that of the big cats: in fact their diet
is similar to that of the Felidae, although they do not always kill their prey.
(It has been established, however, that the larger spotted hyena is a
powerful group hunter.) In addition to the range of teeth seen in Felidae,
hyenas also have highly developed conical teeth (the penultimate pre-
molar, both upper and lower), which they use as hammers to crush the
bones of large herbivores. When the predators and the majority of carrion
eaters have finished eating, the tougher bones still remain: these enclose a
very fatty, and therefore energy-rich, material, the marrow. Hyenas can
gain access to this resource using their specialized premolars. At a certain
point, some other animals also began to compete for the marrow con-
tained in the strong-walled bones of the large herbivores; these were
primates – humans, who, lacking the appropriate teeth, used stone tools
to split the bones.

Arsuaga / Chosen Species 1405115327-4-009 Final Proof page 133 4.6.2005 2:39pm

(carnassial tooth)


Canine Incisors




Figure 9.1 Jawbones of (top) a lion; (below) a sheep

Arsuaga / Chosen Species 1405115327-4-009 Final Proof page 134 4.6.2005 2:39pm

Teeth, Guts, Hands, Brain

But not all land-based carnivores have such a specialized diet, and
therefore dentition, as the cats. The Mustelidae (the family which includes
badgers, otters, martens, mink, and weasels), the Canidae (the group of
the foxes, wolves, coyotes, African wild dogs, etc.), the Ursidae (bears),
and the Viverridae (such as genet and mongoose) have molars behind the
carnassial teeth with chewing surfaces instead of cutting edges. While the
front part of the lower carnassial teeth has a cutting edge, the back does
not. This indicates that their diet also includes vegetable matter which is
more difficult to absorb, and requires prior chewing, insalivation, and
predigestion in the mouth. The famous cave bear, a gigantic animal bigger
than any modern bear, which lived at the same time as Neanderthal and
Cro-Magnon man, was basically vegetarian, as its large molars indicate.
It goes without saying that the large herbivores also modified their
teeth to the function they had to perform. The great grasslands – the
steppes, prairies, and pampas – provide enormous quantities of plant food
which is, however, of low nutritional value because the plants which make
it up are rich in fibers like cellulose. In order to absorb the food, a high
level of predigestion in the mouth is required, where the fibers are
reduced to a pulp which is then processed in the digestive tract. This
processing requires the assistance of symbiotic microorganisms (basically
bacteria) which live in the intestines of herbivores and can perform the
miracle of decomposing cellulose into carbohydrates which can be
absorbed by the animal.
The teeth of these mammals have a ridged crown which increases the
efficiency of their chewing surfaces. The work of maceration and reducing
the vegetable matter to a pulp is speeded up if the surfaces rubbing
against one another are ridged, like millstones in traditional grain mills
(Figure 9.1).
The premolars and molars of herbivores have to withstand intensive
wear, since the fibrous stalks of plants like grasses also contain mineral
particles which make them harder and more abrasive. They therefore have
very high crowns, so that they last longer, and very wide chewing surfaces;
they have no cutting edge for slicing flesh.

The Teeth of Primates

What kind of teeth do our closest relatives, the great apes, have? And are
ours similar to those of herbivores, or carnivores, or like those of the great
apes? The latter have very large incisors and canines, particularly the males.

Arsuaga / Chosen Species 1405115327-4-009 Final Proof page 135 4.6.2005 2:39pm

Teeth, Guts, Hands, Brain

This development of the canines is common to all anthropoids, particu-

larly in males, and it does not indicate an adaptation to a carnivorous diet.
The canines are used for fighting, both within the species and against
predators. Baboons, geladas, and mandrills are well known for the males’
enormous canines, which leads even successful predators on anthropoids,
such as leopards or even lions, to treat them with respect. However, they
cannot be considered in any way carnivorous, although they do on occa-
sion catch insects, small vertebrates, and baby mammals.
The premolars of the Old World monkeys and the apes have two cusps
(like those of humans), but the lower anterior premolar is long and
narrow, and the outer cusp is so much higher than the inner that it can
be considered to have only one cusp. The posterior edge of the upper
canine slides against the anterior edge of the first lower premolar, creating
a scissor-like action. This type of anterior premolar with a cutting action is
known as sectorial. There is also usually a space between the upper lateral
incisor and the upper canine, into which the lower canine inserts when the
mouth is closed; this is known as a diastema.
The molars are the most interesting teeth in terms of inferring type of
diet. There is a fundamental difference in the crowns of the molars
between the Old World monkeys and the apes. In Old World monkeys
each molar has two transverse ridges. These monkeys have a wide range of
diets, but as we have seen, the ridged form of the teeth is typical of
mammals which eat plant products requiring a degree of predigestion in
the mouth. In particular, the group of the colobus monkeys and langurs,
which eat large amounts of leaves, have teeth with higher cusps and
sharper ridges.
In apes the cusps, four on the upper molars and five on the lower
molars, form isolated projections separated by dips (without the transverse
ridges seen in the Old World monkeys). To see what this actually looks
like, you can look at your own teeth in a mirror (assuming that the relief of
the crowns has not been altered by fillings) (Figure 2.1). This shape is not
specialized in any particular way, making it fairly well adapted to the
general type of diet of apes, which includes fruit and also the less tender
parts of plants. Gorillas especially eat the tougher parts of plants, since
they have become too large to climb trees looking for fruit: in line with
their fiber-rich diet, gorillas have higher cusps on their molars.
The great apes in general, and particularly chimpanzees, have very large
upper central incisors, with a straight cutting edge so that they function
like chisels. The upper lateral incisors are smaller, the lower incisors even
smaller, but all have a straight cutting edge and the same chisel function.
It is very difficult to eat an apple using only your molars and premolars.

Arsuaga / Chosen Species 1405115327-4-009 Final Proof page 136 4.6.2005 2:39pm

Teeth, Guts, Hands, Brain

The front teeth (incisors and canines) work to slice up fruits of a certain
size so that they can then be chewed by the back teeth (premolars and
We modern humans have modified the dentition of our ancestors
considerably. To begin with, our incisors are not as large as those of the
apes. We have no diastema, and the tips of the canines project hardly at all
beyond the chewing surfaces of the other teeth; moreover, with wear the
point of the canines very quickly becomes level with the surface of
the adjacent teeth. In humans and apes the incisors are spatulate, or fairly
flat, widening from the base of the crown (or neck) to the cutting edge. In
apes the canines are conical, but in humans their shape is closer to that of
the incisors, spatulate, and becoming wider until just below the tip of the
tooth (which is in any case fairly rounded).
In humans the first lower premolar has two cusps like the others, and it
does not slide against the posterior edge of the upper canine. In apes the
incisors are far in front of the canines, which are aligned with the pre-
molars and molars, so that the dental arches have parallel sides and are U-
shaped. In humans the dental arches are parabolic or elliptical.

The Teeth of the First Hominids

Of all the fossil hominids, the one which is most similar to living apes is
Ardipithecus ramidus. In this species the canines project considerably,
although not as much as in apes, and the first lower premolar is sectorial.
Other typical features of apes which have become lost over the course of
human evolution are found in this hominid, including the morphology of
the dental arch, which is U-shaped with parallel sides. No palates have
been preserved, but judging from the size of the canines we can imagine
that there was a well-developed upper diastema.
In Ardipithecus ramidus the first milk molar has a very prominent
principal cusp, as in apes; this characteristic had never before been seen
in a hominid (all the others have several cusps of similar size). This trait is
so significant that a small fragment of jawbone showing a conical first milk
molar appeared on the cover of the historic issue of Nature in which, on
September 22, 1994, the discovery of Ardipithecus ramidus was pub-
We have already noted that the enamel is also very thin in this species.
We have to conclude that the diet of Ardipithecus ramidus differed little
from that of modern chimpanzees. However, there is one trait which

Arsuaga / Chosen Species 1405115327-4-009 Final Proof page 137 4.6.2005 2:39pm

Teeth, Guts, Hands, Brain

merits further attention. Although the canines project much further than
in modern humans, they do not project as much as in apes, nor are they
the same shape. As we have mentioned, in humans the canines are spatu-
late or incisiform (having the shape of the incisors), and this condition
appears already incipient in Ardipithecus ramidus. Moreover, the anterior
edge of the first lower premolar is not sharpened through wear against the
upper canine (as it is in apes). In fact, this change in the function of
the upper canine/first lower premolar pair, small though it is, is one
of the main arguments for classifying Ardipithecus ramidus as a hominid.
This combination of traits prompted Tim White and his colleagues to
attribute a crucial role in human evolution to this species, as ancestor of all
the subsequent hominids, and thus our first known ancestor. Given its
antiquity (4.4 million years), it is unlikely that much more primitive forms
of hominid will be found.
Although when the first fossils of Australopithecus afarensis were discov-
ered they seemed the height of primitivism (as had Australopithecus afri-
canus, long before), the archaic traits of Ardipithecus ramidus give an idea
of how many things had changed in Austalopithecus afarensis. Some of the
fossils of the latter still show first lower premolars with a very prominent
cusp (while others already have two cusps), the diastemas in the upper
dental arches are small or nonexistent, and the canines project less and are
more incisiform than in Ardipithecus ramidus. The most important feature
is that in Australopithecus afarensis the molars have become larger, and
particularly wider, than in Ardipithecus ramidus. This, combined with the
increase in thickness of the tooth enamel, indicates that a major ecological
change had taken place, resulting in a diet which is still plant-based, but
now contains a large proportion of tough, abrasive foods. There is no trait
in the dental morphology which indicates significant consumption of flesh.
Our first ancestors, both those who lived in the humid forest like Ardi-
pithecus ramidus and those who had begun to exploit the resources of dry
woods and clearings like Australopithecus afarensis, were not hunters but
vegetarians. To put it more dramatically, the first hominids did not come
down from the trees to become ‘‘murdering apes,’’ nor was it their taste for
meat that prompted them to abandon the forest.
The fossil record for Australopithecus anamensis is still scarce, but we
have some information on the masticatory apparatus of these fossils which
are intermediate between Ardipithecus ramidus and Australopithecus afar-
ensis. The Kanapoi maxilla and mandible, together with the teeth that
have been preserved, show very primitive characteristics, but it appears
that the enamel had already become thicker and the size of the molars
increased to a certain extent.

Arsuaga / Chosen Species 1405115327-4-009 Final Proof page 138 4.6.2005 2:39pm

Teeth, Guts, Hands, Brain

The traits which distinguish Australopithecus afarensis from Ardipithe-

cus ramidus are still more marked in Australopithecus africanus; for ex-
ample, in the latter all the first lower premolars have two cusps.

Size of the Molars and Shape of the Hand

One characteristic of the masticatory apparatus that we have so far men-

tioned only in passing is closely related to the type of diet: this is the size of
the chewing surface. This variable also depends on the size of the indi-
vidual, so we need to separate the two factors (body size and type of diet).
If a species shows a large difference in size between the sexes, the males
will have larger teeth than the females – not only the canines, which are
not directly related to diet, but also the premolars and molars, which
are. This is due simply to the fact that having a larger body to maintain
means needing to process more food, or to increase the amount of
chewing done. This is why gorillas have bigger teeth than chimpanzees.
But gorillas’ teeth also have more work to do than those of chimpanzees
because gorillas include in their diet plant products which are more
fibrous and less nutritious than the fruits which chimpanzees eat. Scien-
tists therefore have to work out how to establish how much of the
difference in tooth size between the two species is due to difference in
body size, and how much is due to the type of diet.
Clearly this is a similar problem to that of relative brain size which we
have already discussed, and it is resolved in the same way. We can construct
an index of megadonty, or tooth size ratio, between the actual size of the
molars in a given fossil species (this value forms the numerator), and that
which should correspond to a primate of the same size (the denominator).
The latter, hypothetical value is calculated using an equation which relates
the size of the molars to body size for a set of modern species. We can also
approach the problem more directly: an average female Australopithecus
afarensis, which would weigh a little less than an average female chimpan-
zee, should have slightly smaller molars than the female chimpanzee if her
diet was the same. However, those of the fossil species are larger, indicat-
ing a different diet and more laborious chewing in Australopithecus afar-
ensis. The molars of Ardipithecus ramidus, on the other hand, are smaller
than those of Australopithecus afarensis, and comparable to those of
chimpanzees, as their diet must also have been.
Using the method described above, Henry McHenry has calculated an
index of megadonty (he measures the chewing surface for the set of teeth

Arsuaga / Chosen Species 1405115327-4-009 Final Proof page 139 4.6.2005 2:39pm

Teeth, Guts, Hands, Brain

formed by the last premolar and the first two molars). The value of the
index for Australopithecus afarensis is almost double that for chimpanzees,
indicating that diet had already changed to include tougher plants. The
relative size of the chewing surface increases somewhat in Australopithecus
africanus and continues to increase in Paranthropus robustus and Paran-
thropus boisei (to three times that of chimpanzees). However, in Homo
habilis the value of the index returns to the level of Australopithecus
afarensis, and in Homo ergaster it approaches that of the chimpanzee,
very similar to our own.
Paranthropus are the hominids with by far the largest chewing surface,
both in absolute terms and in relation to body mass. And not only the
molars, but also the premolars are larger, particularly the posterior pre-
molar, which has become ‘‘molariform,’’ increasing the number of cusps
and acquiring the form of a molar. The incisors and canines, on the other
hand, have become very small, creating an imbalance between the very
reduced anterior teeth (incisors and canines) and the enlarged posterior
teeth (premolars and molars).
Large canines which project beyond the level of the adjacent teeth and
have to be slotted into diastemas make it difficult to move the front part of
the jaw from side to side. This type of lateral movement combines with the
vertical movements of opening and closing the mouth to produce rota-
tion. Rotational movements are necessary to grind tough plant foods. The
great apes do not have this problem because their diet does not require so
much grinding. In baboons the face is so large that some rotation of the
posterior part of the jaw is possible, although in males the anterior part is
blocked by the hugely developed canines. In Australopithecus, Paranthro-
pus, and Homo, the wear on the premolars and molars is very intensive; as
soon as these teeth appear their chewing surfaces characteristically become
flat and with virtually no relief, the enamel disappearing rapidly and
wearing closer and closer to the dentin, the tissue beneath the enamel.
What kind of diet did Paranthropus have? No doubt similar to that of
modern baboons, based on hard grass seeds, fresh or dry legumes, and nuts
(Figure 9.2). It would also have included fruits like blackberries, berries, and
stone fruit, which were easier to chew, and the underground storage organs
of plants like bulbs, tubers, fleshy rhizomes, and tuberous and swelling roots
(baboons eat practically everything, including small animals).
We have already noted that grass seeds and legumes are rich in carbo-
hydrates (which they store principally in the form of starch), and are thus
very nutritious, although low in protein (legume proteins, moreover, have
to be cooked in order to be absorbed). As processing seeds in the mouth is
very laborious, Paranthropus developed their own personal mill, through

Arsuaga / Chosen Species 1405115327-4-009 Final Proof
page 140 4.6.2005 2:39pm

Figure 9.2 Paranthropus boisei and Theropithecus brumpti. These two primates developed somewhat similar specializations in their
chewing apparatus, owing to a similar diet, containing a high proportion of tough plant matter
Arsuaga / Chosen Species 1405115327-4-009 Final Proof page 141 4.6.2005 2:39pm

Teeth, Guts, Hands, Brain

the expansion of the premolars and molars and of the muscles which
moved them and the bones in which the muscles were anchored. This
powerful chewing apparatus would also have been useful in cracking nuts,
rich in vegetable oils. The underground storage organs of some plants
provide carbohydrates and some proteins, although the soil which they
would inevitably have ingested with them would have contributed to wear
on the teeth. Baboons, and particularly their relatives the geladas which
live on the high, virtually treeless plains of Ethiopia, can digest grasses and
grass stems, but this is a unique specialization which cannot be assumed
without evidence to have existed in hominids.
Some authors believe that Paranthropus, and also the australopithec-
ines, used sticks to dig bulbs or tubers, for example, out of the earth. It is
also possible that both genera included animal products in their diet, as do
baboons and chimpanzees when they can. However, the theory main-
tained by Dart and Le Gros Clark, that the first hominids were predators
on a large scale, using tools to kill their prey, is now completely discre-
dited. In a magnificent example of taphonomic analysis, Bob Brain dem-
onstrated that the hominids whose bodies accumulated in the caves of
South Africa were not the fierce occupants of the caves, but rather prey
brought there by the real hunters, leopards. The baboons, gazelles, and
other herbivores whose remains are found together with the hominids
would have met the same fate. Lee Berger and Ron Clarke have recently
postulated that even the Taung Child was the victim of a predator,
although this time a winged one – an eagle.
One interesting aspect of baboons’ diet is that it requires a certain
capacity for manipulation of small objects, such as seeds, with the hand.
There is no published data on what Ardipithecus ramidus’ hands were like,
but it is possible that in the first hominids (this species or an earlier one)
the morphology of the hand was similar to that of apes, with a short
thumb and long palm and fingers. Remember that this is an adaptation for
hanging from trees by forming a hook with the hand.
In Australopithecus afarensis the hand is already almost the same as
ours, although the phalanges are more curved, somewhat like those of
chimpanzees; moreover, the thumb is proportionately shorter and the rest
of the hand slightly longer than in our own species. Hand bones of a
modern type have been found in the Swartkrans deposit. Most of the
Swartkrans fossils are from Paranthropus robustus, but as there are also
some from Homo, we cannot be sure to whom these hand bones belong.
However, we can venture the hypothesis that, at least since Australopith-
ecus afarensis, hominids had modified their hands so as to have full
capacity for manipulation of small objects, as a result of their new diet.

Arsuaga / Chosen Species 1405115327-4-009 Final Proof page 142 4.6.2005 2:39pm

Teeth, Guts, Hands, Brain

Our vision of the first hominids has changed greatly with time. Origin-
ally they were envisaged as fierce predators of the savanna, and then as
inoffensive vegetarians, also living in the savanna. Ardipithecus ramidus is
now believed to have lived in the humid forests, and all species of Aus-
tralopithecus are thought to have been vegetarians living in a mixed
environment, whether this is called dry forest and clearing or densely
wooded savanna. Two types of hominids which were genuinely specialized
in open environments, Paranthropus and the first Homo, seem only to
have appeared 2.5 million years ago.

Guts and Brain

In 1891 Sir Arthur Keith made an observation which passed unnoticed at

the time. He had noted that in primates there was an inverse relationship
between brain size and stomach size. Surprisingly, the larger the stomach,
the smaller the brain – in other words, a primate cannot have a large
digestive system at the same time as a large brain. This fact urgently
required explanation. However, we had to wait for this explanation for
over a century until, in 1995, Leslie Aiello and Peter Wheeler put forward
a hypothesis which is extremely important for studies of human evolution.
Aiello and Wheeler point out that, since the brain is one of the most
costly organs in terms of metabolism (the economy of the body), an
increase in the volume of the brain would only be possible if it was
balanced by a reduction in the size of some other organ with a similar
energy consumption. In proportion to their mass, the organs with the
highest energy consumption in the human body are the heart, the kid-
neys, the brain, and the ensemble formed by the digestive tract and the
liver. The brain accounts for 16 percent of the organism’s basal metabolic
rate (the energy consumption, per unit time, required to maintain the vital
functions of an individual at rest); the digestive tract represents a similar
proportion, at 15 percent. We have already seen, in the section on ence-
phalization, that human beings have a brain substantially bigger than that
which corresponds to a hypothetical nonhuman primate of our size; it
turns out that our digestive tract is smaller than that which would be
expected by almost exactly the same proportion.
Aiello and Wheeler conclude that the enlargement of the brain which
occurred in Homo was only possible with a shortening of the digestive
tract. The length of the digestive tract depends on the kind of food it has
to process: in carnivores it is always shorter than in herbivores because

Arsuaga / Chosen Species 1405115327-4-009 Final Proof page 143 4.6.2005 2:39pm

Teeth, Guts, Hands, Brain

meat is easy to digest. Herbivores, on the other hand, need long digestive
tracts in order to metabolize the plants they eat, particularly if these are
rich in cellulose.
We have already noted that about 2.5 million years ago two types of
hominid settled in open environments. This is a point in time which many
authors consider important from the point of view of climate change,
because a general cooling of the planet occurs – reflected in Africa by the
definitive expansion, at the cost of the enclosed forest environments, of
the great grasslands and the savannas (albeit still more or less wooded).
One of these two types of hominid is Paranthropus, which adapts its
chewing apparatus to eat the tough but nutritious plants of the savanna,
as baboons do today.
However, the brain of Paranthropus does not increase in size as much as
that of Homo. Bearing in mind that this increase in brain size implies an
increased energy consumption, only two solutions remain. One is increas-
ing the basal metabolic rate of the organism as a whole (the total energy
consumption). This is not what happened, since humans have the rate
which corresponds to a mammal of our size. The other solution is to
reduce the energy consumption of another organ in order to balance the
body’s energy economy. The question is, which organ will have to reduce
in size? Not the heart, nor the kidneys, nor the liver, which are vital parts.
However, the digestive tract can become smaller if diet is improved, by
increasing the proportion of high-quality nutrients, those which can be
easily digested and have high calorific value. What are these high-quality
products which do not form part of the diet of Paranthropus? The only
answer can be animal fats and proteins. The first humans must have shifted
to incorporating a larger proportion of meat than any other primate in
their diet; they would have got this first as carrion-eaters, and then
increasingly as hunters.
For the first time in the history of mammals, this change in diet was not
reflected in a change in dental morphology. We do not find humans with
teeth that work like hammers to crush bones, nor with teeth that act like
knives to cut meat, because the tools required for splitting bones and
cutting skin and meat are outside the body: they consist of stones and the
edges of stones worked by humans.
Thus, the enlargement of the brain in Homo could only have occurred
with a change in diet; this in its turn is reflected in the reduction in size of
the digestive tract and a correlative reduction in the chewing apparatus.
Aiello and Wheeler insist that this does not mean that change in diet
automatically leads to an increase in brain size; they maintain only that we
had to become carnivores in order to become intelligent (although this is

Arsuaga / Chosen Species 1405115327-4-009 Final Proof page 144 4.6.2005 2:39pm

Teeth, Guts, Hands, Brain

a vicious circle, since locating high-quality food requires greater mental

The last Paranthropus disappeared in Africa, where they emerged,
about one million years ago. Perhaps their ecological niche was continu-
ally being reduced through competition with baboons and geladas and
with humans, our ancestors. The curious aspect of this is that, since the
development of agriculture, most of humanity has sustained itself largely
on plant products which, although cultivated, are fairly similar in com-
position to those eaten by Paranthropus. The difference is that we do not
grind the hard seeds of cereals and legumes with our teeth, or crack nuts
with them; since Neolithic times we have been cooking the seeds or
making them into flour using artificial grinders. We learned to crack
nuts with a stone long before this.

Arsuaga / Chosen Species 1405115327-4-010 Final Proof page 145 4.6.2005 2:40pm



Yes. One goes on. And the time, too, goes on – till one perceives ahead a
shadow-line warning one that the region of early youth, too, must be
left behind.
Joseph Conrad, The Shadow Line

The Rhythm of the Molars

Broadly speaking, the four main stages of life in catarrhine primates

(which include ourselves) are marked by the eruption of the three molars
of our adult dentition. The emergence of the first molar marks the end of
what we might call early childhood, a period of strong dependence on the
mother, from whom the child is rarely separated, and which basically
covers the period of nursing (only in humans does it include several
more years of development).
The appearance of the second adult molar coincides with the end of the
second period of childhood, and the beginning of the important changes of
puberty; in males an increase in the level of testosterone in the blood is
detected shortly afterward.
The emergence of the third molar, known as the wisdom tooth, cor-
responds to the completion of development and the beginning of adult
life, although the final fusion (knitting) of all the bones occurs somewhat
While the three major periods of development are essentially the same
in all primates, their duration varies. In humans development continues

Arsuaga / Chosen Species 1405115327-4-010 Final Proof page 146 4.6.2005 2:40pm


over about twenty years, almost double that of the great apes (chimpan-
zees, gorillas, and orangutans). On average chimpanzees get their first
adult molar when they are a little over three years old; in our species it
erupts at the age of six. In chimpanzees the second molar emerges at the
age of six and a half, while in humans it appears at around eleven years;
finally, the third molar appears in chimpanzees at around age eleven, and
in ourselves at around eighteen (although among modern humans the
four wisdom teeth sometimes do not appear at all, and when they do it can
be well into adulthood). In chimpanzees the first period of estrus, with its
characteristic swelling of the ano-genital region, occurs between the ages
of nine and almost fourteen, depending on the nutritional condition of
the female (in humans too the first menstrual period, or menarche, occurs
earlier, on average, in girls who are well fed and healthy). In chimpanzees
as in humans, this is usually followed by a period of infertility before the
first conception. In chimpanzees the first birth occurs on average at the
age of fourteen. A chimpanzee can live to the age of forty or more.
Despite this difference in length of the periods of development between
apes and ourselves, the relatively close correspondence between the stages of
life and the appearance of the molars helps us to establish the state of
development of a fossil hominid: although we do not know its ‘‘chrono-
logical age’’ (i.e., how old it was when it died), we do know its ‘‘physiological
age’’ (whether it was an infant, an older child, an adolescent, or an adult).
Thus the famous Taung Child was emerging from early childhood
when the eagle snatched him, because the first molar was erupting pre-
cisely at that moment. So how old would he have been? If he died between
the ages of three and four this would mean that the rhythm of develop-
ment of Australopithecus africanus was similar to that of chimpanzees,
while if he died at six, it would be slower, like our own. When Turkana
Boy died the second molars had already erupted. Was he around eleven
years old, like a modern child in terms of the development of his skeleton
and teeth, or was he around seven years old, like a chimpanzee with the
same ‘‘physiological age’’? How can we find out? In order to tackle this
question, we shall start at the beginning of existence – birth.

Birth and the Newborn

Shortly before birth the fetus shifts into an inverted position, with the
head in the upper part of the mother’s pelvis (formed by the iliac crests
and known as the greater pelvis or false pelvis). During birth, the full-term

Arsuaga / Chosen Species 1405115327-4-010 Final Proof page 147 4.6.2005 2:40pm


fetus has to pass through the lower part of the pelvis (known as the lesser
or true pelvis) via a bony tube called the pelvic cavity or birth canal.
Among chimpanzees, gorillas, and orangutans birth is easy and rapid,
because the birth canal is large in relation to the head of the fetus. In
these apes the entry to the birth canal (known as the pelvic inlet) is oval,
with its major diameter running from front to back (a sagittal or ante-
roposterior orientation, in anatomical terms), and its minor diameter
oriented transversely (from side to side). As we have seen, the modifica-
tion in pelvic architecture required to make it possible to walk upright
meant that the joints between the coxal bone and the spine, and the coxal
bone and the femur, drew closer together. As a result, the sagittal diameter
of the birth canal reduced, and this is why complications can occur in the
human birth process.
In human females the entry to the birth canal is not oval but round
(Figure 10.1). The major diameter of the pelvic inlet is not sagittal: it is
sometimes transverse but more often neither sagittal nor transverse, but
on the two oblique or diagonal diameters. Moreover, the head of the full-
term fetus is elongated, so that its major diameter is anteroposterior (from
the forehead to the nape of the neck). The position of the fetus’ head
adapts to the greater diameter of the pelvic inlet, generally one of the
oblique diameters. As the greatest dimension of the outlet to the birth
canal is always sagittal (in humans as well as in other primates), the skull
and shoulders of the human fetus have to enter the birth canal with one
orientation (transverse or oblique) and leave with another (sagittal). Thus
there is a rotation within the birth canal both of the head and, afterwards,
of the shoulders (Figure 10.2).
To complicate matters further, in humans the vagina is angled forward,
forming a right angle with the uterus, so that as it passes through the birth
canal the fetus moves not in a straight line, but around a pronounced
curve which ends immediately below the pubic bone, where the baby’s
head emerges. In order to accommodate this curved trajectory, the fetus’
spinal column arches, the head flexes strongly toward the back and the
crown of the head faces forward as it is born, so the baby is facing
backward (in the opposite direction to the mother). In other primates
the vagina lies at the same angle as the uterus, with which it is aligned, and
the full-term fetus follows a straight trajectory toward the back as it is
born; furthermore the face is forward (in the same direction as the
mother). To sum up, birth in humans is ventral, while in other primates
it is dorsal.
Karen Rosenberg and Wenda Trevathan have pointed out that the
mother ape can help her baby to be born, guiding it through birth with

Arsuaga / Chosen Species 1405115327-4-010 Final Proof page 148 4.6.2005 2:40pm






Iliac crest

Acetabulum inlet



Figure 10.1 Morphology of the pelvis in a chimpanzee and a human female.

The arrow indicates the direction taken by the full-term fetus during birth

Arsuaga / Chosen Species 1405115327-4-010 Final Proof page 149 4.6.2005 2:40pm


Figure 10.2 Flexion and torsion of the fetus during birth. To begin with
(top), the head lies in a transverse or diagonal position and the shoulders in a
sagittal position (along the mid-plane of the body). Later (center) the head lies
sagittally and the shoulders transverse or diagonally. As it comes out the head
flexes strongly backwards. Finally (below) the head, once it has emerged, shifts
transversely, and the shoulders turn to a sagittal position to come out

Arsuaga / Chosen Species 1405115327-4-010 Final Proof page 150 4.6.2005 2:40pm


her hands, clearing its nose and mouth so that it can breathe better and
freeing it from the umbilical cord if this is wound around its neck. Birth in
primates is a solitary event, with no outside help. However, in humans the
mother cannot see the newborn’s face because it faces in the opposite
direction, and any attempt to pull it could, given the extreme dorsal
flexion of the head, damage the spinal cord. Because of this women in
all cultures seek assistance at the moment of birth; human birth is a social
activity rather than a solitary behavior.
We have two australopithecine pelvises in a reasonably good state of
preservation: that of Lucy, and that of Sts 14 (an Australopithecus africa-
nus from Sterkfontein). Both belong to small individuals, and are there-
fore probably female. In Lucy’s pelvis the entrance to the birth canal is
very wide in the transverse direction. This could be a trait of the species or
simply an individual character, since in Sts 14 the entrance to the birth
canal is more rounded (in some women the birth canal is very wide from
side to side, a condition known as platypellic). In any case, in Lucy the
fetus’ head would have lain transversely, accommodating itself to the form
of the pelvic inlet. Robert Tague and Owen Lovejoy believe that the
position of the head would be the same on leaving the birth canal (there
would thus be no rotation), and that the fetus’ trajectory would be linear
and toward the back, rather than curved and toward the front (Figure
10.1). So this would be a transverse type of birth, a form without known
equivalent among primate species.
However, the morphology of the ischium and the pubis leads us, like
Christine Berge, to conclude that the vagina was angled forward, not
backward, in female australopithecines; this would mean that the birth
process in australopithecines would be similar to that of modern humans,
with rotation and a curved trajectory.
Before moving on in our discussion of birth in early hominids, let us
spend a few moments considering the implications if this theory were
proved true. In primates the vagina opens dorsally (toward the back) and
penetration during copulation is from behind. Gibbons and orangutans,
which spend much time in the trees, often mate hanging from their arms
and face to face, but chimpanzees and gorillas copulate from behind. The
bonobo has very varied positions, including face to face. In humans this is
the position favored by the ventral (forward) position of the vulva, and it is
characteristically human. We are also distinct in this, and if our conclusion
about the direction of the vagina in australopithecines is correct, face-to-
face copulation would also be the rule among these primitive hominids.
There has been much discussion of whether birth for Lucy would be as
difficult as it is for modern women, or as simple as it is for the great apes.

Arsuaga / Chosen Species 1405115327-4-010 Final Proof page 151 4.6.2005 2:40pm


In fact the latter are the exception to the rule among primates: in gibbons
and other catarrhine and platyrrhine monkeys the diameters of the head of
the fetus are fairly close to those of the birth canal. This gives rise to a
certain number of pelvic–cephalic conflicts which result in the death of the
fetus, the mother, or both. This is an unfavorable consequence of the fact
that anthropoids have large brains in proportion to their generally small
body size.
In women the pubis is longer (in relation to the overall size of the
pelvis) than in men. As the pubic bone forms the anterior wall of the birth
canal, this lengthening of the pubic bone in women produces a wider
birth canal. Lucy and Sts 14 have an extraordinarily long pubis, and thus a
much larger birth canal, in relation to body size, than modern women.
The extremely elongated pubis in Lucy and Sts 14 gives rise to a pelvic
morphology known as hyperfeminine according to modern standards.
These and other characteristics lead us to conclude that the pelvis of
australopithecines was already designed to resolve the problem of birth
that bipedalism had created. Although this problem was perhaps not as
acute as in our species, the advantages enjoyed by female chimpanzees,
gorillas, and orangutans had been lost.
We can consider this question of the rhythms of development right
from the moment of birth, since among apes the newborn comes into the
world with a brain more than one-third of the volume of the adult brain,
while in our species it is less than one-quarter. This means that at birth the
newborn human is much less developed cerebrally, and therefore more
helpless, than any other primate. Primates in general are a group of
mammals which give birth to few offspring at a time (usually only one),
but these are well developed when they are born. This developmental
precocity is also found among apes, so the less developed state of our
newborns must be a condition our evolutionary line has acquired since the
time of any ancestors whose babies were more developed.
We have no direct information as to the size of the newborn australo-
pithecine’s brain, since no neonate skull has been found, but we can
approach the problem in another way. The pelvises of Lucy and Sts 14
tell us the dimensions of the birth canal through which the head and
shoulders of the fetus would have to pass during birth. The dimensions of
the birth canal establish the maximum size of the skull and the brain of the
newborn; in the case of Sts 14 and Lucy this would be comparable to that
of a newborn chimpanzee, or perhaps a little bigger. Furthermore, the
brain size of the average adult Australopithecus afarensis (Lucy’s species)
and Australopithecus africanus (the species of Sts 14) would also be
approximately the same size as that of a chimpanzee (about 400–410 g),

Arsuaga / Chosen Species 1405115327-4-010 Final Proof page 152 4.6.2005 2:40pm


or perhaps bigger. Thus, the ratio between the brain size of the newborn
and that of the adult (which indicates the degree of cerebral development
of the newborn) would be comparable in the great apes and in the first
hominids. So the newborn babies of the first hominids would have come
into the world much more developed than modern human babies, al-
though the birth process was already like that of modern humans because
the australopithecines were bipedal like us.
We have already seen that in Homo ergaster body size had increased
substantially in relation to australopithecines, Paranthropus, and Homo
habilis. But at the same time the brain volume of the adult doubled.
The brain of the newborn may also have doubled in size, leaving the
proportion between the two brain sizes the same. But it is also possible
that the dimensions of the birth canal did not increase by the same
amount; in this case the brain of the fetus could not be as large (since
it had to pass through the birth canal), and it would represent a smaller
percentage of the adult brain than in australopithecines. This would
imply that the newborn came into the world less developed, needing
more care.
Although this supposition accords with the generally held view (which
we shall discuss later) that Homo ergaster already exhibited major changes
in the social sphere compared to australopithecines, and we imagine a
more protective social environment, we have no definitive proof that
young Homo ergaster were born less mature. Unfortunately, the only
pelvis of the species we have is that of the Turkana Boy, which in addition
to being that of a child is very incomplete and allows us to draw no firm
conclusions on this subject. Nevertheless, it is worth mentioning that
Christopher Ruff and Alan Walker have estimated, on the basis of the
Turkana Boy, that the newborn of the species would have a brain one-
quarter the size of the adult brain. Bearing in mind that any such conclu-
sion is as yet unproven, this would nevertheless imply that the newborn
Homo ergaster was already less mature.

Childhood and Adolescence

Having established that the newborn australopithecine would have been

no more developed than newborn great apes of today, we need to deter-
mine how long the different life stages lasted in the first hominids.
When he published Man-apes or Ape-men in 1967, Wilfrid Le
Gros Clark thought that the development of australopithecines and

Arsuaga / Chosen Species 1405115327-4-010 Final Proof page 153 4.6.2005 2:40pm


Paranthropus was similar to that of modern humans, slower and of longer

duration than that of the apes.
Fortunately, we have a way of establishing the age at death in some
cases, based on analysis of the lines of growth of the tooth enamel. Enamel
is deposited in layers as the crowns of the teeth form; if we cut a cross-
section through the enamel, the boundaries between the layers can be
distinguished as what are known as striae of Retzius, clearly visible under a
microscope. Each of these lines corresponds to the end of a cycle of
deposition of enamel lasting approximately one week. By counting the
number of striae of Retzius in a tooth where the development of the
crown is not complete, or was completed shortly before the individual’s
death, we can work out how many weeks elapsed from when the tooth
began to form. If we know when the formation process began, we can
determine how old the individual was when he died. We do not need to
cut through a fossil tooth to observe the striae of Retzius, because they
can be seen on the surface.
This method of calculation has been applied to the incisors of some
hominids in which the crown is newly formed or has recently finished
growing. There is a fossil of Australopithecus africanus (Sts 24, from the
Sterkfontein deposit), and another of Australopithecus afarensis (L.H.2,
from Laetoli), which show a stage of dental development similar to that of
the Taung Child. Remember that at the beginning of this chapter we
established the premise that if the Taung Child died between the ages of
three and four, we would conclude that development in his species (Aus-
tralopithecus africanus) was like that of chimpanzees. In fact, when the
method of lines of enamel deposition was applied to incisors of Sts 24 and
L.H.2, the ages of death obtained were between 3.2 and 4 years. Tim
Bromage and Chris Dean, who carried out this investigation, also obtain
ages close to those of chimpanzees for the emergence of the first molars in
Paranthropus. Everything thus seems to indicate that Le Gros Clark was
mistaken and that the life stages of australopithecines and Paranthropus
were no different in duration from those of the great apes.
The method of counting the lines of enamel cannot be used on the
Turkana Boy’s incisors, because he was already too old when he died, and
we thus cannot say with certainty at what age he died. If we assume
development patterns similar to those of modern humans, he would
have died at the age of 11, while if his development was like that of
chimpanzees and gorillas, he would have been about 7 years old.
Fortunately, there is an indirect (although less precise) way of approach-
ing the problem. It has been observed that in all primates brain size is very
closely correlated with longevity and with the duration of the different life

Arsuaga / Chosen Species 1405115327-4-010 Final Proof page 154 4.6.2005 2:40pm


stages. The greater the average brain of a species, the longer the life of the
individuals of that species and, for example, the later the second adult
molar appears. The brain volume of Homo ergaster is thought to have
averaged between 800 and 900 cc, midway between that of chimpanzees
(and australopithecines and Paranthropus) and our own. On this basis we
can, in principle, assume that the rhythm of the development of the
species would also be roughly midway between the two, and that Turkana
Boy died at the age of nine or ten. If this is the case, it would mean that a
significant change in the biology of development had occurred in this
species compared with australopithecines and Paranthropus.
There is one further aspect we need to consider in order to complete
this look at development. Turkana Boy is estimated to have been about
160 cm tall. However tall he might be when fully grown, this height
would be too tall for a modern child at his stage of development (as we
have noted, equivalent to about 11 years old). In other words, the height
of Turkana Boy would be more suited to a modern adolescent of 15 or
older (even among very tall populations such as the Maasai).
However, in our species an acceleration in growth occurs at an age older
than that of Turkana Boy – the so-called ‘‘growth spurt’’ of puberty
(around the age of twelve in girls and fourteen in boys). The conclusion
drawn by Holly Smith, who has studied the development of Turkana Boy,
is that in Homo ergaster this change in the rate of growth did not yet exist,
and that it only occurs in our species. In apes and other primates devel-
opment is more continuous, without any marked accelerations in adoles-
cence, and the percentage of growth remaining is lower in apes of an
equivalent ‘‘physiological age.’’ No doubt this was also true of Homo
ergaster (though even so, Turkana Boy would have grown to be over
180 cm tall).

Arsuaga / Chosen Species 1405115327-4-011 Final Proof page 155 4.6.2005 2:41pm


Social Intelligence

As long, therefore, as we followed these reflections, we could not but

conclude that man is by nature fitted to govern all creatures, except his
Xenophon, Ciropaedia

The Unexciting Sex Life of the Female Orangutan

Five years after her last period of sexual activity, the female orangutan
comes into season once more. During this long break she has been
gestating, giving birth to and nursing her latest offspring. Now the
moment has come for weaning and beginning a new cycle, a new gesta-
tion. An ovum awaits fertilization, probably by the same male as the
previous time. After this brief receptive period, the sex life of the female
orangutan ceases for another five years or so (unless the baby she has
conceived miscarries or dies).
Female chimpanzees and gorillas also have seasons (known in zoology
as periods of estrus) separated by breaks of several years, generally more
than three years and less than six, but while female gorillas have sexual
relations with only one male, female chimpanzees will have several lovers.
Human women, on the other hand, do not have a period of estrus, and
thus the moment of ovulation cannot be detected; unlike female apes,
their sexuality is not governed by being in season and it includes the long
infertile periods of gestation, lactation, and menopause. In other words,
whereas female sexuality in apes is linked exclusively to reproduction, in
our species it also exists beyond this function.

Arsuaga / Chosen Species 1405115327-4-011 Final Proof page 156 4.6.2005 2:41pm

Social Intelligence

What causes the differences in sexual behavior, and by extension in

social behavior, in species that are so closely related? The answer is
genes – the same factor that causes male orangutans and gorillas to be
much more heavily built than females, and the males of these species
to compete amongst themselves to form harems. In chimpanzees and
humans there is less difference in size between the sexes, and in neither
case do the males fight one another to establish groups of females
(see Figure 2.3). Just as some genes determine our physical characteristics,
others program our behavior, and both kinds are subject to the harsh test
of natural selection over the course of the generations.

Behavior as Adaptation

With their research, Konrad Lorenz and Niko Tinbergen (who shared the
Nobel Prize for Medicine with Karl von Frisch in 1973) laid the founda-
tions for understanding animal behavior, creating a new scientific discip-
line – ethology, or the study of behavior. Ethology establishes that there is a
degree of genetic programming which codetermines behavior (in other
words, it determines it only in part). Just like morphological structure and
physiological function, behavior has to adapt to the way of life of the
individuals of different species, and therefore the genes which determine it
are subject to natural selection.
Ethologists have demonstrated that many of the behavior patterns of
animal species are innate, and moreover that they develop, like the organs,
over the life course of individuals. Thus young animals have certain infantile
behaviors which are only useful for surviving at this stage of life, when they
are strongly dependent on their parents, in a situation of direct competition
for food with other young. Complex behaviors of courtship, mating, and
care of offspring, however, mature at the same time as the reproductive
organs, and in many cases develop even if the animal has grown up in
extreme isolation, reflecting their innate rather than learned nature.
Konrad Lorenz (1903–89) discovered that the geese he studied at his
family home in Altenberg were programmed to recognize as mother the
first moving object the goslings saw at the moment when they hatched. In
normal conditions this would be the true biological mother, but they can
be made to adopt as mother a person or even an inanimate object which
appears in front of them at the decisive moment. Thus the different signals
which trigger behaviors in animals can be broken down and analyzed,
making ethology an experimental science.

Arsuaga / Chosen Species 1405115327-4-011 Final Proof page 157 4.6.2005 2:41pm

Social Intelligence

Perhaps the reader is surprised or disappointed to find that there are

genes underlying our behavior, just as there are for the color of our eyes,
or belonging to one or other sex. But what science requires is not acts of
faith, but experimental proof of hypotheses, and today there is no doubt
that there is an element of genetic determinism in behavior. In any case,
would it be preferable if human beings came into the world as a ‘‘blank
slate,’’ with nothing written on it? If learning processes were the only
thing which determined our behavior, would it not be much more terrible
to be in the hands of those who have the power to program education?
How could we be free if we are entirely conditioned by the education we
have received?
Of course, ethology does not oblige us to take the reductionist course of
believing that all our behavior is planned, from the cradle to the grave, and
that we have no capacity to make our own decisions. In fact, programming
of this kind would not be very adaptable because each individual lives in his
or her own ecological and social environment, and has to adapt to it. An ant
is much more strictly programmed in its behavior patterns than a mammal.
Humans are an extremely intelligent species of social primate, and we have
great flexibility in our behavior, allowing us to respond differently, on
the basis of our own experience or education, to the different situations
we face in our environment. Life throws up many unpredictable problems,
and the solution cannot therefore be in the genes.
We shall return to this subject at the end of this chapter, because one
decisive factor in the expansion of our brain seems to have been the need
to analyze and take decisions in relation to a particularly changeable and
unpredictable aspect of our environment – the behavior of other members
of our group. As social primates living in large communities, we need to
process a large quantity of information on an enormously complex system
(the community) that comprises many elements (individuals) relating to
one another in a virtually infinite number of ways.
With the development of information technology in today’s world, it is
easy to understand that the more instructions a computer has, the more
software that is loaded into it, the greater its flexibility and capacity to do
different things – including being more efficient at analyzing situations
and taking decisions. Very soon computers will even be able to learn from
their own experience. In other words, genetic programming is not an
enemy of freedom, but allows us to evaluate the different options and
choose between them.
At the beginning of this book we noted that the way an individual
uses or does not use its organs during its own life has no effect on the
organs of its descendants (despite what Lamarck maintained). Similarly,

Arsuaga / Chosen Species 1405115327-4-011 Final Proof page 158 4.6.2005 2:41pm

Social Intelligence

the information accumulated by an individual about his environment

during his life cannot be transmitted through the genes. However, this
wealth of useful knowledge does not necessarily have to be lost, since it
can be transmitted between generations by non-genetic methods,
through education. In the case of humans, the passing of information
from one generation to another is called culture; this type of permanent
collective memory is partly universal and partly varies with each ethnic
group, each community, each family. Because it is cumulative, it has made
possible the great advances in science and technology. Lorenz’s goslings
had no genetic programming to tell them exactly what their mother
looked like, but they had some simple rules telling them how to find
out for themselves. Similarly, we humans have an innate disposition for
learning a language from a young age, but our genes do not program us to
learn English, Spanish, or Chinese.
Having made this foray into the principles of ethology, let us return to
the social life of our closest relatives and attempt to enter into this aspect
of biology which does not fossilize. In this area we owe a great deal to the
pioneering work done by Jane Goodall with chimpanzees, Dian Fossey
(1932–85) with gorillas, and Biruté Galdikas with orangutans.

Comparative Sociobiology of Hominoids

All anthropoids (Old and New World monkeys, apes, and humans) are
social, with one exception. The term ‘‘social species’’ is used to refer to
those in which lasting relationships between adults are established. The
only exception to this rule are the orangutans, which are solitary animals
in that the only stable bonds are those between mothers and their non-
adult offspring. Adult males and females only come together during the
brief and very widely spaced periods of estrus of the females. There are no
relationships or alliances between males or among females. Each female
lives in her small territory with her offspring, and the wider territories of
the males encompass those of several females, with whom, however, they
only come into contact to reproduce. Males compete with one another for
territory and for the females within it, and this competition results in them
becoming much heavier, twice the weight of females. It could be said that
male orangutans form harems, but these are harems in which the females
live dispersed rather than all together.
Gorillas, on the other hand, are highly social. As their food supply is
abundant and constantly available they do not need large territories and

Arsuaga / Chosen Species 1405115327-4-011 Final Proof page 159 4.6.2005 2:41pm

Social Intelligence

they do not travel great distances in one day. Each group of gorillas is
formed by an adult male, the silverback, and his harem, a group of females
with their offspring, all descendants of the silverback. When a female or a
male reaches puberty they leave the group. Males compete with one
another for females, and this is why they are so big, although the differ-
ence between the sexes (known as sexual dimorphism) is less marked than
in orangutans: the average female gorilla weighs about 60 percent of the
average male, or a little less than two-thirds.
Common chimpanzees do not form harems. When females reach adult-
hood they generally leave the group; however, when males reach
adulthood they stay in the group. This means that all the males in a
group of common chimpanzees are related, while the adult females are
not. Each community controls a territory, which the males defend against
other alliances of males. These struggles between groups of related males
are violent fights, sometimes to the death.
The trees which supply the fruit eaten by chimpanzees are dispersed,
and the fruit ripens at different times; for this reason within each territory
there are moments of fusion, when many individuals gather around a tree
with ripe fruit, and of fission, or dispersal in search of sources of less
abundant resources. When a female is in season, marked by a spectacular
anogenital tumescence (swelling of the zone around the sex organs), there
is no competition among males; rather, several of them mate with the
female at different times. Because of this promiscuity and the absence of
harems the males are not much bigger than the females, who on average
weigh at least 80 percent, or almost four-fifths, of the average male.
In discussing the behavior of apes we cannot emphasize too strongly that
we are talking about primates, not insects; thus their rules of behavior are
very varied and any attempt to generalize them is a blatant simplification.
For example, sometimes male chimpanzees at the top of the hierarchy have
been observed to push in front of those lower down the hierarchy when
mating with females in estrus; there are also cases where male–female
couples form, lasting several days, or even the full two weeks of estrus.
Genetic analyses of the relationship between individuals in chimpanzee
communities are currently underway; in the future these will throw a great
deal of light on how all of these behavioral strategies are reflected in
reproductive success. In order to make these ‘‘paternity tests,’’ hairs
which have fallen out in the ‘‘nests’’ the chimpanzees build in the trees
to sleep, and mouth cells left on bitten fruit, are used. On the basis of
this method, Paul Gagneux and his colleagues have observed that females
frequently ‘‘escape’’ from their community and are fertilized by a male
from another group.

Arsuaga / Chosen Species 1405115327-4-011 Final Proof page 160 4.6.2005 2:41pm

Social Intelligence

In pygmy chimpanzees (bonobos), the bonds between related males are

not as strong, but there are closer relationships between females, even
though they are not genetically related. Curiously, these interfemale rela-
tionships prevent the isolated males from imposing their hierarchical
authority. On the other hand, the males do not appear to take defense
of the community’s territory very seriously. Sexual dimorphism is similar
to that of the common chimpanzee.

Natural Selection and Sexual Selection

When Darwin spoke of natural selection he was referring to the elimin-

ation of defective individuals and the survival of the best adapted. Spon-
taneous variations produce new types of organism, often unviable, but
sometimes with characteristics which allow them either to cope better in
their way of life (to occupy their ecological niche better, as we would say
now), or to exploit resources that their competitors do not use (expand-
ing or changing their ecological niche). This is how new species appear.
However, Darwin also realized that in many species the males showed
characters which were not adaptive from the ecological point of view.
These characters make the males more showy, or stronger (as we have
seen in the case of gorillas and orangutans), sometimes giving them
weapons to fight other males of their own species (and only secondarily
useful in struggles with animals of other species). In order to explain this
apparent exception to his theory of natural selection, Darwin developed
the theory of sexual selection. Briefly, this means that females select the
most decorated male, or passively accept the one who defeats all the other
males, showing in either case that he is an individual with excellent health
and strength, and thus the best possible progenitor among those com-
peting. Darwin put forward this theory in his book The Descent of Man,
and Selection in Relation to Sex (1871). As he put it, sexual selection
‘‘depends on the advantage which certain individuals have over others of
the same sex and species solely in respect of reproduction.’’ However,
Darwin could not have imagined that competition was established at a
level below that of the individual – at the level of the sperm.
The great primatologist Adolph Schutz (1891–1976) observed in 1938
that primates vary greatly in the size of their testicles relative to body weight.
For example, a male chimpanzee weighing 45 kg has testicles weighing
approximately 120 g (60 g each), while a male gorilla of 160 kg has testicles
which together weigh 30 g. Schultz did not know at the time how to

Arsuaga / Chosen Species 1405115327-4-011 Final Proof page 161 4.6.2005 2:41pm

Social Intelligence

interpret this difference, but the modern anthropologist Alexander Harcourt

and his collaborators have put forward a very original hypothesis.
Comparing the relative weight of the testicles in different genera of
primates, it can be observed that the species in which the males have very
large testicles are those in which the social groups include several males,
such as chimpanzees, baboons, and macaques. Gorillas, on the other
hand, which live in groups where there is only one male, have relatively
small testicles. The size of the testicles appears to be related to the quantity
of sperm they produce, and also to the length of the tail and the mobility
of the sperm. Harcourt’s hypothesis is that in species with large testicles, a
female may be inseminated by several males when she is in season, and the
sperm of the males competes, in quantity and quality, to fertilize the egg.
Male gorillas, however, do not have to compete in terms of sperm,
because they do not let any other adult male approach their group of
females. Gibbons, which are strictly monogamous, orangutans, which are
polygamous but not social, and humans have a testicle weight no higher
than normal for a primate of their size.
In the social categorization of primates, humans are usually described as
monogamous – a definition which might cause some readers to shake
their heads! Admittedly there is wide variation in the human family
structure in different cultures (which have little to do with the social
context in which our evolution took place). It is obvious that gibbons
are monogamous and do not form groups, that orangutans are polygam-
ous and live separately, that gorillas are polygamous and form harems, and
that chimpanzees are social and promiscuous, but surprisingly, we are not
as sure about the social biology of our own species.
From the biological point of view no specific answer can be given.
However, one thing is certain: humans form social groups which include
many male individuals, so it would seem appropriate to place us in
the group of primates in which there is selection among sperm, like the
chimpanzees. In strictly zoological terms, the fact that this is not the case
means that we are a species in which it is rare for a female to have sexual
relations with several males during the period of ovulation. Moreover, as
ovulation is not signaled overtly as it is in other primate species, males have
no way of knowing when it occurs, so we can simplify the previous sentence
to say that it is rare for a female to have sexual relations with several men.
Furthermore, the level of sexual dimorphism in the body weight of our
species (around 83 percent) indicates that there is not a high degree of
competition for females among the males.
In other words, we are a special type of primate, and we see male
individuals living together in society, but also a degree of exclusivity in

Arsuaga / Chosen Species 1405115327-4-011 Final Proof page 162 4.6.2005 2:41pm

Social Intelligence

the sexual relations of each man with a woman, at least for a period of
time. In other words, something like monogamy.
But having set out what we know and what we do not know about the
social biology of our species and other species of primate, it is time to
investigate how the first hominids behaved in society.

Bipedal and Monogamous from the Beginning?

Rob Foley observes that there are no examples among apes of groups
based on related females (whereas this is common among Cercopitheci-
dae, or Old World monkeys: only in the red colobus monkey are there
relationships between related males). He concludes that the social struc-
ture found among chimpanzees, involving alliances between related males
to defend a common territory, and dispersal of adult females outside their
birth territory, must also have been the rule among the first hominids,
whose social biology would initially have differed little from that of
However, australopithecines and especially Paranthropus lived in more
arid environments than modern chimpanzees, in sparser woods where the
sources of food would have been more dispersed and less abundant. This
leads Foley to believe that australopthecines and Paranthropus would have
had larger territories than chimpanzees. They would probably have main-
tained alliances between related males, both for defense of resources
against other coalitions of males, and against predators, which would
have become more dangerous as the tree cover decreased. However,
smaller social units would have formed within these large territories,
since the resources available in the environment would not have allowed
all members of a given community to be together at all times. Thus, the
social system would also have been one involving fusion, with much of the
group gathered around a very abundant source of food, to travel long
distances over open ground, or to sleep, and fission, with the group
separating into smaller units to find food during the day.
Determining what these smaller units might have consisted of is an-
other question. Jane Goodall and her colleagues have studied the spatial
distribution of a population of just under 150 chimpanzees in Gombe
National Park, in Tanzania, since 1960. The most closely studied com-
munity in the park was made up, over the years, of between 4 and 13 adult
males, 8 to 18 adult females and 18 to 31 immature individuals. The
group’s territory ranged from 6.75 to 15 sq km, occupying between

Arsuaga / Chosen Species 1405115327-4-011 Final Proof page 163 4.6.2005 2:41pm

Social Intelligence

3 and 6 valleys in the central region of the park. The adult females spend
more than 65 percent of their time alone with their young, feeding in
their own small areas of around 2 sq km, which partially overlapped with
those of others. The adult males are more social, and travel throughout
the community’s territory, patrolling and defending their borders to-
gether. This type of spatial distribution, with females spending time
alone with their young, is unlikely for australopithecines, even more so
for Paranthropus, because of the risk from predators in a more open
However, the family units might have been comparable to those of
hamadryas baboons (Papio hamadryas) and geladas, i.e., formed by one
male with a number of females (and their young); these may occasionally
have joined together with other small harems to form large groups to find
food, travel, or sleep. This is a social system which we know to function
successfully in ecosystems similar to those of our ancestors, like the arid
environments where the hamadryas baboons live. But one author, Owen
Lovejoy, believes that australopithecines were monogamous, and more-
over that monogamy is closely related to the development of the hom-
inids’ bipedal posture. Let us look at his argument.
Monogamy, or pairing off for reproduction, is not found only in
humans. It is also observed in many birds and primates, including gib-
bons. But as we have already noted, our species has the additional unusual
feature of a permanent sexual relationship, for most of the time without
reproductive function. To put it even more clearly, defining human sexu-
ality purely in terms of procreation is not natural (in the biological sense),
but the absolute opposite. Among humans sex also exists to keep the
couple together; in other words, it is at the service of love.
However, this romantic function of sex does not contradict Darwinist
principles, but in fact reinforces them. The long period of development of
human young makes it impossible for a mother to look after several
offspring at once (in the context of a hunter-gatherer economy). The
stable couple, in monogamy, means that the father is involved in the task
of supporting the family, which operates as an economic as well as a
reproductive unit.
There has been much debate as to what is understood by the contribu-
tion of fathers (males) to the raising of young, how this contribution can
be measured, and to what extent it occurs in the different primates and
human societies. Whatever the case, the situation in our species bears no
relation to the habits of chimpanzees, gorillas, and orangutans (our closest
relatives in other senses), where males have no connection with their
offspring; all we can say is that among chimpanzees and gorillas (not

Arsuaga / Chosen Species 1405115327-4-011 Final Proof page 164 4.6.2005 2:41pm

Social Intelligence

orangutans, which are not social), the males are tolerant of the young and
protect them against predators and being killed by other males (a serious
risk, as we shall see).
Lovejoy maintains that bipedalism has nothing to do with our ances-
tors’ adaptation to open environments, as has so often been suggested,
and that it probably arose when we still lived in the forest. Bipedalism,
according to Lovejoy, was not related to temperature regulation, to
efficiency of locomotion, or to freeing the hands to make tools. On the
contrary, bipedalism would have freed our hands and arms to carry food.
In Lovejoy’s view males transported food in this way to the base camp to
feed the females and their young, who thus avoided many dangers by not
having to travel with their mothers.
Lovejoy even believes that this made it possible to reduce the long
period between births characteristic of chimpanzees, gorillas, and orangu-
tans, and to increase the number of young over the fertile life of the
females. This supposed advantage of the first hominids is highly debat-
able, because even in modern human hunter-gatherer societies (i.e., pre-
agriculture and pre-animal husbandry) the interval between births is still
long, from three to four years.
We are used to seeing natural selection from the point of view of
individuals, but in order to understand what follows we need to look at
it from the point of view of genes. As Richard Dawkins noted in his
famous book The Selfish Gene, individuals die, but genes remain. Or
more correctly, they are preserved as copies in other bodies, those of our
children. Dawkins takes his argument to the extreme, affirming that genes
use us for their own benefit and even sacrifice our bodies if necessary, as in
the case of what is known as altruistic behavior, when parents put their
own lives at risk to save their children. In fact, the time and energy
expended in raising offspring, although less heroic, can also be considered
altruistic, since it is not aimed at satisfying the interests of the parents, but
those of the children. Everything has a point according to the logic of
natural selection, because the genes of parents who do not behave altru-
istically and abandon their children (thus dispatching them to their death)
will not be present in the succeeding generation.
These explanations enable us to understand behaviors which appear mon-
strous if we view them in moral terms. We noted earlier that when they reach
puberty, female gorillas and chimpanzees leave their native territory and their
community to enter another which is completely alien to them, and into
which they are nevertheless accepted. However, we are talking here of
females without young, because if they appear with a nursing infant in the
alien group it is virtually certain that the baby will be killed. The dominant

Arsuaga / Chosen Species 1405115327-4-011 Final Proof page 165 4.6.2005 2:41pm

Social Intelligence

male in the case of gorillas, or the related males in the case of chimpanzees,
have no interest in the genes carried by the newly arrived young individual,
but they are very interested in its mother being available as soon as possible to
be fertilized. Interrupting lactation by the rapid method of infanticide places
the mother in a situation where she can begin a new ovarian cycle. The reader
can imagine what happens to the young in a harem of gorillas when the
silverback is replaced by another male (as a result of natural death or defeat in
a fight). The logic of the genes is implacable.
If we follow this logic, a male of the first bipedal hominids providing
food for a female with young, as Lovejoy suggests, would have to be sure
that the young carried his genes. If the females of the species had periods
of estrus they would have to be watched closely during the whole of this
period. On the other hand, if the female had no estrus, making it impos-
sible to know when she was ovulating (and thus monopolize her during
this period), the only viable alternative to ensure paternity was monogamy
and sexual fidelity.
In fact menstruation may have evolved as an indicator of fertility,
because a few days after it has occurred there is a fertilizable egg. How-
ever, Beverly Strassmann points out that other primates do not appear to
make these calculations, since chimpanzees (like macaques, Old World
monkeys, and baboons), which menstruate copiously, indicate ovulation
to the males by means of swelling of the sexual organs.
As we pointed out above, social life is not preserved in the fossil record.
But we also noted that among gorillas and orangutans, males compete
with one another for females, and that as a result there are great differ-
ences in body size between the two sexes, whereas the two sexes are more
similar in chimpanzees and humans (although there are still differences; in
gibbons, which are strictly monogamous, there is no sexual dimorphism
of weight). If this is a universal rule linking anatomy to social biology, it
could offer a key to approaching the problem in extinct species. This type
of exercise, which consists of applying relationships observed in the mod-
ern biosphere to fossil species (with the premise that such relationships
have always existed), is known as actualism, and is one of the tools most
frequently used to study life in the past.
Australopithecus afarensis, the oldest hominid species of which there is
sufficient fossil record to study the difference between the sexes, shows
great variation. So much that some authors saw two or more distinct
species among these fossils; it was Tim White who grouped all the
fossils into a single species, as we have already noted. In fact this is a
species which shows great sexual dimorphism of size between males and
females (around 66 percent of body weight, very close to that of gorillas),

Arsuaga / Chosen Species 1405115327-4-011 Final Proof page 166 4.6.2005 2:41pm

Social Intelligence

as Henry McHenry, Charles Lockwood, Brian Richmond, William Jun-

gers, and William Kimbel, among others, have highlighted (Figure 11.1).
From this point of view it does not seem that Lovejoy’s hypothesis, that
bipedalism was linked to monogamy from the beginning, is tenable (if this
were the case we would expect much less marked sexual dimorphism in
Australopithecus afarensis). A social system closer to that of hamadryas
baboons and geladas, as suggested by Rob Foley, appears more likely for
the first hominids (Figure 11.2).
But there is one small problem with this interpretation. As Michael
Plavcan and Carel van Schaik have shown, in modern primates the length
of the crown of the canine is an even better indicator than body weight of
the level of competition among males for females. Now, as a result of the
reduction of the canines in Australopithecus afarensis, the difference in the
length of the canines between males and females is less than that in gorillas
and orangutans, and also less than in both species of chimpanzee: it is still
smaller in ourselves.
If male Australopithecus afarensis competed for females, why would
they have had to give up their best weapons? The reduction of the canines
may be related to the change in diet experienced by these hominids, since
large canines impede lateral movements of the jaw. Moreover, in view of
the intense wear to which the canines of australopithecines and Paran-
thropus are immediately subject, we might wonder what would be the
point of having large canines which would barely project at all, precisely at
a time when they would be more useful as weapons. To sum up, the
absence of major sexual dimorphism in the canines among the first hom-
inids appears to have been imposed by the type of chewing, rather than
indicating a lack of competition among males for females.
Nevertheless, according to Tim White the beginnings of reduction of
the canines are already observed in Ardipithecus ramidus, which is not
thought to have had a very different diet from chimpanzees, although it
might have been sufficiently different to explain this reduction. In fact,
this combination of major sexual dimorphism in body size combined with
very little sexual dimorphism in the size of the canines, characteristic of
hominids at least since Australopithecus afarensis, is not found in any
modern primate species, and comparison is therefore impossible.
Thus we have no conclusive answer to the question of when modern
patterns of social biology appeared in human evolution. Current thinking,
which tends to consider australopithecines and Paranthropus as ‘‘bipedal
chimpanzees,’’ vegetarians with no stone tools, no major increase in brain
size, no language, and a short period of maturation, suggests that this
great change took place within the genus Homo.

Arsuaga / Chosen Species 1405115327-4-011 Final Proof page 167 4.6.2005 2:41pm

Social Intelligence

Figure 11.1 Male and female Australopithecus afarensis

Arsuaga / Chosen Species 1405115327-4-011 Final Proof page 168 4.6.2005 2:41pm

Social Intelligence

Figure 11.2 One male and two female Australopithecus afarensis

Arsuaga / Chosen Species 1405115327-4-011 Final Proof page 169 4.6.2005 2:42pm

Social Intelligence

Brain Size and Size of Social Group

In a previous chapter we investigated the use of being bipedal, or what

kind of adaptation this is and what ecological niche it relates to. We have
already seen that there is no easy answer to this question. However, no
one wonders what the use of being intelligent is. We are so convinced that
intelligence is a gift which makes us superior to any other living form that
we do not concern ourselves with its adaptive value. Nevertheless, en-
largement of the brain is a specialization like that of any other organ, and
natural selection favored it because it offered advantages in the context of
the ecological niche occupied by hominids, in those in which it occurred
(not all of them, as we have seen). What were these advantages?
There are two moments in human evolution where a marked increase in
brain size occurred which could be related to significant changes in social
patterns. The first of these increases occurred with Homo ergaster, in
whom brain volume increases from approximately one-third of the aver-
age in modern humans, as in australopithecines and Paranthropus, to two-
thirds (Homo habilis had a brain size somewhere in between). The second
great enlargement occurs during the last half-million years, and produces
the enormous brains of our species and the Neanderthals.
As we have seen, the increase in brain volume implies a change in diet,
because it affects a tissue with high energy consumption. As a result,
animal proteins and fats become incorporated into the diet in substantial
quantities. Unlike some of the very abundant (but much less energy-rich)
plants, these resources are not continuously distributed in the environ-
ment, and are not easy to obtain; thus the size of the territory to be
covered and the time spent searching for them increase. At the same time,
from Homo ergaster onward rates of growth of young are already close to
our own, implying a longer period of dependence than in apes and earlier
hominids. All of this means that it would be difficult for a mother alone to
take care of a number of young at the same time. It is therefore possible
that the great social change took place in Homo ergaster, although some
authors maintain that it took place during the second great cerebral
expansion, that of humans and Neanderthals.
But there is more than just an indirect relation between increase in brain
size and relations between the sexes: it is possible that brain enlargement is
directly related to an increase in social complexity. In the first place, it has
been observed that mammals which live in complex societies (such as
anthropoids and dolphins) have larger brains than solitary mammals of
similar size. Aiello and Dunbar have also discovered that among the

Arsuaga / Chosen Species 1405115327-4-011 Final Proof page 170 4.6.2005 2:42pm

Social Intelligence

different species of primates, the relative size of the neocortex in relation

to the rest of the brain is directly related to the size of the social groups
formed by these species.
However, no such relationship has been identified between relative
neocortex size and way of life, making the ‘‘ecological theory’’ of the
origin of our highly developed intelligence less attractive than the ‘‘social
theory.’’ Nevertheless, we must keep in mind the fact that from the first
Homo, hominids entered an ecological niche which was entirely new for
primates, that of carrion eaters and hunters; this may also have favored
intellectual development.
To sum up, the theory that expansion of the brain and increased
intelligence (or at least a substantial part of this expansion) represents an
adaptation to social life, an environment in which individuals have to both
cooperate and compete with other individuals, is very plausible. An intel-
ligence developed on this basis (‘‘social intelligence’’) could easily be
applied to other types of complex situation. In order to do well in this
difficult social environment various tactics need to be used, from the
formation of alliances with other individuals, based on genetic relationship
or interest, to deception. Anne Pusey, Jennifer Williams, and Jane Goodall
have observed that even among chimpanzees it is important to be born
into a ‘‘good family.’’ Although female chimpanzees often gather food
alone, and do not seem to have a marked hierarchy among themselves, the
offspring of high-ranking females have better life prospects than those of
other females, probably because they have access to better sources of food.
The skills required for what Andrew Whiten and Richard Byrne have
called ‘‘Machiavellian intelligence’’ (in reference to the fifteenth-century
writer Niccolò Machiavelli, whose book The Prince gives advice on how to
triumph in politics through the use of hypocrisy and unscrupulous means)
include, of course, a good memory, in order to remember exceptionally
complex social organograms (who is who). In addition, at least among
humans, there is a certain capacity for intuiting the intentions of another
and anticipating his actions, as well as the capacity for creating mental
representations of hypothetical situations (not only remembering past
situations), evaluating them, and acting on the result – in other words, a
capacity for thinking.

Arsuaga / Chosen Species 1405115327-4-Summary Final Proof page 171 8.6.2005 5:23pm


Here ends the part of human evolution that took place exclusively in the
African continent – our birthplace, and the place where we literally took
our first steps. It is time to pause for a moment to look back, and sum up
what we have discussed and draw the main conclusions from it.
Darwinism, in the updated form of the modern synthesis, is the tool
which evolutionists of the end of the second millennium use in their
analyses. Although this is not a closed, dogmatic system and it is open to
criticism, improvement, and amendment, it forms the basis for our current
understanding of the phenomenon of evolution. One important aspect of
Darwinism is the adaptive nature of its basic mechanism, natural selection,
which does not appear to impose any specific direction on evolution. But in
this book we are describing, on the basis of the fossil record, how human
evolution arose; not until the final chapter will we discuss whether this has
any meaning, whether it corresponds to some Law of Evolution.
After setting out the terms of the problem (what happened, how it
happened, and why it happened), we have opted to follow the advice of
Chilo (usually attributed to Socrates): ‘‘Know thyself.’’ We are primates,
more specifically anthropoids, and as such we are basically visual, intelli-
gent mammals, diurnal, tropical, living in forests and trees. Many of our
morphological, physiological, and ethological characteristics match this
ecological definition of the group. The fact that humans, and to a lesser
extent other primates, now live in climates, regions, and ecosystems very
different from the backdrop to our evolution is still an anomaly, and a very
recent phenomenon in terms of the long history of the primates. In this
book we have briefly recounted this history, particularly that of the
hominoids, the class of anthropoids to which we belong.

Arsuaga / Chosen Species 1405115327-4-Summary Final Proof page 172 8.6.2005 5:23pm


Hominoids originated in Africa at least 23 million years ago, and they

were numerous and varied in that continent and later also in Asia and
Europe. After reaching their maximum diversity around 10 million years
ago, hominoids began to decline rapidly. Most species had already dis-
appeared by 7 million years ago, and in the present day the only really
abundant and widespread species is our own.
There are two reasons, perhaps related, for the decline of the homin-
oids: the loss of habitat caused by ecological changes, and competition
with the other great group of anthropoids, the Old World monkeys or
Cercopithecidae, which are now more varied and abundant than homin-
oids. The great rain forests and monsoon forests have been retreating as a
result of a global change in climate, with an increasingly marked cooling of
the planet over the last 4 million years. This climate change appears to be
caused essentially by astronomical factors, combined with a particular
arrangement of the continental landmasses. In addition to the change in
climate, there may have been a reduction in carbon dioxide in the atmos-
phere during the last 8 million years, which would have favored the spread
in the equatorial latitudes of the grassy plants which form, for example,
the pastures of the African savanna.
It is nevertheless important to note that the first hominids, our most
remote direct ancestors, did not appear in the expanding grasslands: the
fossils of Ardipithecus ramidus, from 4.4 million years ago, seem to
indicate an entirely forest-based life, with a type of diet similar to that of
modern chimpanzees. It is still not known whether these hominids were
bipedal. Judging from the primitivism of this species, the divergence of
the lines which lead to humans on the one hand, and to the two species of
chimpanzee on the other, must have occurred shortly before, perhaps only
5 or 6 million years ago. This date for the separation of the two lines is
agreed by molecular biologists, who estimate it on the basis of the genetic
difference between species and the inferred rate of change.
Four million years ago there was a different species of hominid living on
the shores of Lake Turkana, known as Australopithecus anamensis. These
hominids walked erect and their dentition indicates that a change in
ecological niche had occurred, and that they incorporated tough plant
products in their diet: in other words, in addition to fruit, leaves, and
tender stems and shoots they also ate hard seeds, nuts, tubers, roots, and
other underground parts of plants. The former type of plant products is
found in the humid forest; the latter are typical of more arid environ-
The hominid fossils from the subsequent million-year period are
found in fairly substantial numbers; these are assigned to the species

Arsuaga / Chosen Species 1405115327-4-Summary Final Proof page 173 8.6.2005 5:23pm


Australopithecus afarensis. The more extensive record of this species al-

lows us to tackle an important question: which of our distinctive features
came first? We have already seen that bipedalism arose before Australo-
pithecus afarensis; however, in terms of encephalization, the intelligence of
these hominids was similar to that of modern chimpanzees. Their period
of growth also appears to have been similar, and young would not have
been born more helpless, requiring more care.
All the fossils cited up to now come from deposits in the Great Rift
Valley, a great fracture through the African continent with a number of
branches. This is because the lake basins which formed along this fissure
provide favorable conditions for the formation of paleontological de-
posits. But in addition, the great mountains and high plains associated
with the Rift Valley make East Africa a zone of lower rainfall than the
center and west of the continent at the same latitude. Because of this
reduced rainfall in East Africa there are no rain forests, but rather sparser
woodland, grasslands with trees and dispersed scrub (the savanna), and
great prairies. It is possible that it was in these more open and arid woods
of East Africa that a primate ecologically very similar to the modern
chimpanzee evolved toward bipedal forms, and toward a diet which
included tough plants. However, hominids soon spread westward from
their place of origin, because fossil remains have been found well into
Chad. These may also be of a different species from Australopithecus
afarensis, which would mean that there had already been an early diversi-
fication of hominids.
Between 3 and 2 million years ago we find various forms of hominids.
One species, called Australopithecus africanus, lived in South Africa dur-
ing the first part of this period, and its fossils are found in cave deposits.
The australopithecines did not live in these caves: their remains were left
there by predators which hunted them.
Around 2.5 million years ago, the group is clearly split into two major
types of hominid. One, Paranthropus, developed a massive masticatory
apparatus, probably a specialization to process tough, abrasive plant food.
Australopithecus afarensis is its more or less direct ancestor. Three species
of Paranthropus are known: the East African Paranthropus aethiopicus,
which is the oldest; Paranthropus robustus, found in several South African
caves; and Paranthropus boisei, an East African form which died out a little
over one million years ago. Australopithecines and Paranthropus showed
marked sexual dimorphism in body size. They may have formed commu-
nities of several related males, each male in his turn gathering a small
harem of females.

Australopithecus bahrelghazali Bahr el Ghazal Hadar Australopithecus afarensis

Middle Awash Ardipithecus ramidus

Australopithecus afarensis
Konso Paranthropus boisei
Homo ergaster

Omo Paranthropus aethiopicus

Paranthropus boisei
Homo habilis

Australopithecus anamensis
Lake Turkana Paranthropus aethiopicus
Paranthopus boisei
Homo habilis/rudolfensis
Homo ergaster
Arsuaga / Chosen Species 1405115327-4-Summary

Olduvai Homo habilis

Paranthropus boisei
Homo ergaster/erectus

Laetoli Australopithecus afarensis

Lake Malawi Homo?

Makapansgat Australopithecus africanus

Swartkrans Paranthropus robustus

Homo ergaster
Australopithecus africanus Taung
Sterkfontein Australopithecus africanus
Paranthropus robustus Kromdraai Homo habilis?
Final Proof page 174 8.6.2005 5:23pm

Figure S.1 Location of the main deposits with fossils of Ardipithecus, Australopithecus, Paranthropus, and the first
representatives of Homo
Arsuaga / Chosen Species 1405115327-4-Summary Final Proof page 175 8.6.2005 5:23pm


Other hominids, the humans, developed their brain and began to

manufacture stone tools. To begin with they were very similar to the
australopithecines, particularly the South African form. There may have
been several forms of these humans which were still similar to the austra-
lopithecines, probably spread over a large part of Africa. It is likely that
Paranthropus and humans evolved in response to the climate change
which resulted in the expansion of open ecosystems. For the first time,
we meet hominoids which can be considered not strictly forest species.
Then, less than 2 million years ago, a species of humans appeared which
was clearly different from all earlier hominids and from their contempor-
aries Paranthropus: this was Homo ergaster. Not only was their brain even
bigger and organized in a different way; in addition, their face had a more
modern appearance, and they were similar in height or perhaps on average
taller than ourselves (the other hominids we have discussed were much
smaller). Their proportions also corresponded to a body structure gener-
ally similar to our own. Their pattern of development was slower than that
of the apes (and other hominids). This longer period of development
implies a more protective social environment, one in which it is possible
for a mother to look after several young at once; probably for the first
time, males became involved in the care and feeding of young.
Two factors are crucial to understanding the enlargement and restruc-
turing of the brain in humans. One of these, a change in diet with the
regular incorporation of animal proteins, made it possible. The other
factor, the increase in social complexity, gave it meaning. Intelligence
developed largely as social intelligence.
These humans were capable of ‘‘inventing’’ a very elaborate stone-tool
technology, bearing witness to their great mental capacity. Finally, they left
Africa and adapted to a wide variety of land and landscapes throughout
Eurasia, as we shall see in Part II.
And now that at last a highly intelligent hominid, more intelligent than
any other primate but less so than ourselves, has emerged, will we see a
linear evolution, a triumphal succession of increasingly intelligent species,
from Homo ergaster to Homo sapiens ? Or, on the contrary, will human
evolution maintain its branching, complex nature, making it impossible to
anticipate the end of the story at any point? Let us move on.

Arsuaga / Chosen Species 1405115327-4-Summary Final Proof page 176 8.6.2005 5:23pm
Arsuaga / Chosen Species 1405115327-4-012 Final Proof page 177 4.6.2005 2:43pm

A New Home
Arsuaga / Chosen Species 1405115327-4-012 Final Proof page 178 4.6.2005 2:43pm
Arsuaga / Chosen Species 1405115327-4-012 Final Proof page 179 4.6.2005 2:43pm


New Locations for Human


Hominid forms as primitive as Pithecanthropus and Sinanthropus

must have lived in Europe or in the Western half of the Old World at
an age still earlier than that in which advanced hominids like those of
Heidelberg and Steinheim appeared.
Franz Weidenreich, The Skull of Sinanthropus pekinensis

Homo erectus and the Settlement of Asia

Since the first humans were neither European nor Asian, but African, at
some point in the past the first human inhabitants of Eurasia must have
made their way there from Africa. When did this happen?
Since 1891, when Eugène Dubois (1858–1941) discovered a calotte (or
skullcap) and femur at Trinil, in Java, the island has supplied a large
number of human fossils. The oldest, over one million years old, are
some remains from the Sangiran region and the calvarium of a child
from Modjokerto (a calvarium is a neurocranium or braincase, a skull
without the skeleton of the face). Recently geochronologists Carl Swisher
and Garniss Curtiss and their colleagues, using the new laser fusion
method of argon-39/argon-40 dating, have dated the Modjokerto child
at 1.8 million years old, and two fragmented skulls from Sangiran (num-
bers 27 and 31) at 1.6 million years old. One partially preserved skull,
Sangiran 4, and some jawbone fragments are probably of the same age.
However, we need to treat these dates with caution, given that the relation
between the fossils and the dated volcanic sediments is not at all clear

Arsuaga / Chosen Species 1405115327-4-012 Final Proof page 180 4.6.2005 2:43pm

New Locations for Human Evolution

(these and many other fossils from Java were collected by amateurs, rather
than being professionally excavated).
Other Javanese fossils, such as the Trinil skullcap, Sangiran 2 (very
similar), and Sangiran 17 (the best preserved skull of any found in Java)
are between 500,000 and a little less than one million years old. Lastly, the
most recent of all the human fossils from Java are the Sambungmacan
calvarium and 14 more or less complete calvaria, two tibias, and pelvic
remains found in the terraces of the Solo river in Ngandong.
Eugène Dubois was a Dutch doctor who enlisted as an army doctor
with the aim of traveling to Java, where he hoped to find the missing link
between ‘‘ape’’ and man, following the theories of the famous evolution-
ary biologist Ernst Haeckel (1834–1919), who believed that we origin-
ated in the forests of Asia, rather than those of Africa. With the skullcap
and femur he found at Trinil, Dubois believed he had discovered the link
he was searching for, and he named it Pithecanthropus erectus (‘‘erect ape-
man’’), known today as Homo erectus. All the Javanese fossils are ascribed
to this species, although the Ngandong remains, the most recent, are
somewhat different, reflecting a degree of evolutionary change.
The Javanese fossils such as Sangiran 4, Sangiran 2, and the Trinil
skullcap itself differ little in their general architecture from the African
fossils of Homo ergaster. Their cranial capacity is estimated at between 800
and 950 cc. In fact the two species share many primitive traits, their
common inheritance from earlier hominids. For example, although the
brain is much larger than in australopithecines, the neurocranium is still
low (or flat), with a very broad forehead, and widest at the base, at the
level of the temporal bone; from here the side walls converge toward the
top, as can clearly be seen from the back.
However, the Javanese fossils show some characteristics not found in
Africans of the species Homo ergaster, and typical of the entire species of
Homo erectus. Essentially the neurocranium is more robust, with thicker
walls, a straight and highly developed brow ridge overhanging the eye
sockets, and another, also very conspicuous, transverse bony reinforce-
ment at the back of the cranium; this is known as the occipital ridge. There
are also some bony thickenings at the top of the cranial vault and in other
areas of the neurocranium. The side view shows that the occipital bone is
sharply angled.
Although there are fossils which do not belong to Homo erectus which
show some of these traits indicating a robust neurocranium (albeit gen-
erally in attenuated form), the combination of all these features is only
found in this species. Moreover, the base of the skull of Homo erectus
shows a series of specializations.

Arsuaga / Chosen Species 1405115327-4-012 Final Proof page 181 4.6.2005 2:43pm

New Locations for Human Evolution

The only fossil specimen from Java with a well-preserved facial skeleton
is Sangiran 17; this has a very broad, fairly flat face, which looks like a
robust version of Homo ergaster fossils such as ER 3733 and WT 15000.
Furthermore, a jawbone fragment and some cranial remains from Java
which have been very sketchily described are extremely robust, so much so
that there are still some authors who seek to create a new genus, Mega-
nthropus, on the basis of these fossils. Nevertheless, until we have more
complete fossil documentation it is too soon to accept the coexistence of
two human species in Java.
From the evolutionary point of view, it seems reasonable to accept that
Homo erectus derives from Homo ergaster, although there are important
changes which justify considering them as separate species. In 1961 Louis
Leakey found an incomplete braincase (OH 9), dated between 1.4 and
1.2 million years old, at Olduvai. From the morphology of what little
remains, this appears to represent an intermediate link between Homo
ergaster and Homo erectus. However, if it is confirmed that Java was first
populated 1.8 million years ago, it would mean that this specimen could
not occupy such a position in human evolution.
In China two presumed stone tools, a jawbone fragment with a molar
and premolar, and an isolated human incisor from the Longgupo deposit,
could be of a similar age to that currently attributed to the first Javanese
fossils. However, the age is debatable, and the jawbone may, according to
some experts, belong to a relative of the orangutans rather than a human
ancestor. The incisor is certainly human, although we do not yet have
enough information on the position of this fossil in the stratigraphy of the
deposit (the sequence of layers which make up the deposit).
Leaving aside the problematic finds of Longgupo, the oldest Chinese
fossil is a very poorly preserved skull from Gongwangling (Lantian), which
appears to be a little less than one million years old. The Chenjiawo
jawbone (also from Lantian) may be of the same age or a little more recent.
The most complete human fossil record in China was found in the
Zhoukoudian cave, about 50 km from Beijing. Since 1921, when excav-
ations began, the remains of many different individuals, including two
braincases, have been found. Unfortunately, almost all the fossils were lost
in 1941 when they were sent, for ‘‘safety,’’ from Beijing to the USA,
under the guard of US marines. The fossils never arrived at their destin-
ation because the convoy was captured by the Japanese, although it does
not appear that they succeeded in seizing the fossils; whatever the case,
nothing more was ever heard of them. As a consolation we have the
magnificent study made by Franz Weidenreich (1873–1948), and casts
of the fossils.

Arsuaga / Chosen Species 1405115327-4-012 Final Proof page 182 4.6.2005 2:43pm

New Locations for Human Evolution

More recent excavations have discovered some new human remains.

The Zhoukoudian fossils were found at various levels of the deposit and
are dated between 250,000 years (the most recent), and 550,000 years
(the oldest). From the evolutionary point of view the Zhoukoudian
human fossils correspond to the same species as those from Java, Homo
erectus. Their cranial capacity, estimated from five calvaria, ranges from
915 to 1,225 cc.

The First Europeans

Recently a human mandible was discovered in Dmanisi (Georgia), in the

southern Caucasus, often called the gates of Europe. This fossil is thought
to be very old, around 1.5 million years, although it may be more recent;
unfortunately it is difficult to tell from a single jawbone what these
hominids were like.
Although, as we have seen, many doubts still remain over the age of the
first Asians, it seems certain that they are well over a million years old.
When did the first humans arrive in Europe?
The oldest evidence of human presence in Europe could be that of
Cueva Victoria, in the Murcia region of Spain. Here the phalange
of a hand has been found, which is certainly that of a primate and could
be human (it appears so to us); alternatively it might be from Theropithecus
oswaldi, a large ape we have already mentioned, which has been
identified in the same cave on the basis of a molar. Unfortunately, the
phalange was picked up away from its original location, and although it
has been assigned to a breccia (layer of coarse-grained rock) dated at more
than one million years old, we can never be completely certain in these
Also in the Iberian peninsula, Josep Gibert and his colleagues have
presented as human some fossil remains from the Venta Micena deposit
in Orce, in the Granada region of Spain. The deposit is a little over one
million years old, but in our opinion the fossils are not human.
For many years after its discovery in 1907, the Mauer mandible (found
near Heidelberg in Germany) was held to be the oldest human fossil in
Europe. Its age is estimated at 500,000–600,000 years. More recently, in
1993, the shaft of a human tibia was discovered in the English deposit of
Boxgrove; this was considered to be of a similar age to the Mauer fossil.
Some authors came to the opinion that the first human settlement of
Europe occurred a little over 500,000 years ago.

Boxgrove Bilzingsleben


La Chaise

Atapuerca Dmanisi
Arsuaga / Chosen Species 1405115327-4-012 Final Proof

Cueva Victoria Ceprano Petralona

Cabezo Gordo Altamura

page 183 4.6.2005 2:43pm

Figure 12.1 Location of the main European deposits of the Lower and Middle Pleistocene
Arsuaga / Chosen Species 1405115327-4-012 Final Proof page 184 4.6.2005 2:43pm

New Locations for Human Evolution

However, the debate on the first human settlement of Europe was to

change irrevocably in the summer of 1994, with the discoveries made at
Gran Dolina, in Spain.

Gran Dolina and the First Europeans

The deposit known as Gran Dolina belongs to the assemblage of karstic

(cave) deposits of the Atapuerca Mountains, very close to the city of
Burgos, which began to be excavated in 1978 under the direction of
Emiliano Aguirre. Since 1991 the project has been led by Aguirre’s
colleagues Juan Luis Arsuaga, José Marı́a Bermúdez de Castro, and
Eudald Carbonell.
In the late 19th and early 20th centuries a railroad was built to carry
mineral ores from the Demanda Mountains to the area around Burgos.
The railroad route traces a great curve, entering and passing through the
Atapuerca Mountains. The trench dug for the construction exposed a
series of caves filled to their roofs with sediment. Among them was the
cave of Gran Dolina, which contains an impressive stratigraphic sequence
18 m thick, with 11 levels numbered from top to bottom.
A deposit is like a book which is read from bottom to top, from the
last chapter to the first. In Gran Dolina the excavation (over an area of
100 sq m) is progressing slowly in the upper levels, the last chapters of the
book. Level 11 (or TD11), which is around 300,000 years old, has already
been excavated, and excavation is about to begin in level 10 (TD10), which
is very rich in fauna and stone tools, and is approximately 400,000 years
old. In addition, samples have been taken from the entire stratigraphic
sequence of the deposit, and a small part of the lower levels has been
excavated, revealing remains of very ancient animals and some stone
tools. The team investigating the Atapuerca deposits therefore knew that
in the lowest levels of the deposit, the first chapters of the book, there were
indications of a very ancient human presence in these mountains. Just as
sometimes we cannot resist the temptation of glancing to the end of the
book to find out how it finishes, the scientist who is excavating feels the
same desire to know the beginning of his story. The way he does this is to
make a sample survey, which is a shortcut to the deepest part of the deposit.
In Gran Dolina a survey of 6 sq m began in 1993, and reached level 6
(TD6) in the campaign of July 1994. During this and the two following
campaigns, about 80 human remains and 200 tools were recovered from
level TD6.

Arsuaga / Chosen Species 1405115327-4-012 Final Proof page 185 4.6.2005 2:43pm

New Locations for Human Evolution

When these human fossils were found their exact age was not known,
but it was clear that they were older than the Mauer mandible. This
conclusion was based on the fauna found in TD6, particularly the pres-
ence of a fossil rodent called Mimomys savini. About 500,000 years ago
this species of rodent gave way to a primitive form of the modern water
rat, known as Arvicola cantiana. The Mauer mandible and the Boxgrove
tibia were found associated with Arvicola cantiana, while TD6 contains
Mimomys savini.
The beginning of the epoch known as the Pleistocene is usually situated
around 1.7 million years ago (for the last 10,000 years we have been in an
interglacial period of the Pleistocene, sometimes considered as a separate
epoch known as the Holocene). The Pleistocene is divided into three
parts, Lower, Middle, and Upper. The boundary between the Lower
and Middle Pleistocene is marked by a change in the earth’s magnetic
field which occurred about 780,000 years ago, when the field changed
from having an ‘‘inverted’’ polarity to the current, ‘‘normal’’ polarity. The
Lower Pleistocene belongs to the chron of ‘‘inverted’’ polarity known as
the Matuyama chron; for the last 780,000 years we have been in the
Brunhes chron of ‘‘normal’’ polarity.
Geological and paleomagnetic studies carried out by Alfredo Pérez-
González and Josep Maria Parés situated the change in magnetic polarity
which marks the boundary between the Lower and Middle Pleistocene
above Gran Dolina level 6 (TD6). Thus, the human fossils and associated
stone tools are more than 780,000 years old. This evidence forces us to
accept that the first settlement of Europe occurred much earlier than we
had thought, although human presence in the continent may not have
been as extensive or as dense as in the last half-million years.
The human fossils in TD6 represent various parts of the skeleton of at
least six individuals who died at different ages. Bearing in mind the
restricted size of the initial survey, it is hoped that when this level is
excavated extensively in a few years’ time, it will provide a very rich sample
of human fossils and tools.
The association of stone tools in TD6, which includes neither handaxes
nor cleavers, is classified as Mode 1. This attribution poses an interesting
problem, because there is Acheulian culture (Mode 2) in East Africa 1.6
million years ago, and in ’Ubeidiya (Israel) shortly afterwards. So why are
the first European fossils not associated with this later culture?
The same problem is posed by the deposits in China and Java, which
similarly have no handaxes or cleavers. There are various possible solu-
tions. Perhaps the first humans to settle in Europe and Asia abandoned
the Acheulian style of carving of their ancestors; it has been suggested that

Arsuaga / Chosen Species 1405115327-4-012 Final Proof page 186 4.6.2005 2:43pm

New Locations for Human Evolution

the populations of the Far East may often have used another raw material
for their tools, such as bamboo.
Perhaps, as Eudald Carbonell and his team suggest, the people of East
Africa who developed the Acheulian industry had a more advanced culture
than those who manufactured Mode 1 tools. As a result, their population
increased and they forced groups who did not have the superior culture to
the marginal areas of the African continent, and finally to migrate out of
Finally, the new dating we mentioned above for Java and China sug-
gests a new argument: perhaps the first Asians left Africa before the
Acheulian culture arose. The same explanation is also possible for Europe,
but here we have no reliable dates of such antiquity for the initial settle-
ment. Not far from Venta Micena, in Fuente Nueva-3, an assemblage of
stone tools (without handaxes or cleavers) somewhat older than those of
Gran Dolina has recently been found, and it is always possible that others,
even older, may appear somewhere in Europe at some time in the future.
In any case, the first European fossils of the Middle Pleistocene, such as
the Mauer and Boxgrove fossils, are associated with Acheulian technology.
We have to wonder how the Acheulian culture came to Europe – whether
it was with a new wave of settlers, or simply that the technology traveled
from Africa, passing from some populations to others without any migra-
tion of peoples or flow of genes. The problem of whether technological
changes observed in a particular place imply the arrival of a new type of
human is a recurrent one in prehistory, and we shall come across it again.

Prehistoric Cannibalism

In theory, the human fossils should not be in the Gran Dolina cave. In fact,
human remains ‘‘should’’ never be found in karst deposits hundreds of
thousands of years old. What ‘‘should’’ be found, however, and frequently
are found, are bones of carnivores and herbivores, and stone tools. The
reason for this is very simple. The carnivores (lions, panthers, lynx, hyenas,
wolves, wild dogs, foxes) settled in caves, making their shelters and lairs
there. They lived, reproduced, and died there. Bears also spent the winter
(hibernated) in caves. Herbivores do not live in caves, but carnivores and
carrion eaters transport their carcasses there to eat them. Human beings
occasionally used caves, spending some time there, usually not very long,
making and using stone tools, and then leaving them there. Humans also
brought animal prey to the deposit, and ate it there. Finally they left.

Arsuaga / Chosen Species 1405115327-4-012 Final Proof page 187 4.6.2005 2:43pm

New Locations for Human Evolution

In order for a human bone to fossilize in a cave something extraordin-

ary would have to have happened. Perhaps a carrion eater, a hyena for
example, might bring in part of a human cadaver or simply a bone found
outside. But for the skeletons of at least six different individuals to
accumulate, circumstances would have to be more exceptional, more
fortunate for ourselves, although perhaps not so much for the humans
in question. Because in this case, we are looking at the remains of a
cannibal feast.
The human remains in Gran Dolina appear mixed with animal remains,
and fairly broken up. Some show cut marks created by the edge of a stone
tool which has been enthusiastically used to separate flesh from bone. It is
thus clear that the bones were stripped of flesh and eaten in the cave itself
by other humans. This is the oldest known evidence for this kind of
practice. It is difficult to imagine that it relates to ritual behavior, and it
appears in principle that human bodies were treated with no more respect
than those of the herbivores with which their remains are mixed. How-
ever, the studies currently being carried out by a number of researchers on
the Atapuerca team will tell us more.

Homo antecessor

When new fossils are found, the paleontologist compares them with those
found earlier in other deposits, in an attempt to determine to which
species they belong. Sometimes the comparison shows that the new fossils
are unlike any others, and at this point a new species is created for them.
This is the procedure which was followed with the human fossils of Gran
Dolina. Following many studies and comparisons, in 1997 José Marı́a
Bermúdez de Castro, Juan Luis Arsuaga, Eudald Carbonell, Antonio
Rosas, Ignacio Martı́nez, and Marina Mosquera created the species
Homo antecessor (antecessor – ‘‘pioneer,’’ the one who comes before
Establishing the position of the new species in human evolution is
another matter. The Gran Dolina fossils show primitive traits in dentition
and other parts of the skeleton, which is to be expected given that they
are 800,000 years old. These archaic characters are no longer found in
later European fossils, and this is why the humans of Gran Dolina are
considered to be of a different species from fossils such as the Mauer
mandible, which is approximately 500,000 years old. On the other hand,
the species represented in Gran Dolina is not Homo erectus, since it lacks

Arsuaga / Chosen Species 1405115327-4-012 Final Proof page 188 4.6.2005 2:43pm

New Locations for Human Evolution

the specializations of the latter species. Finally, these first European hu-
mans could represent a late population of Homo ergaster. However, there
are a number of indications that this is not the case.
An adolescent jawbone fragment from Gran Dolina is less robust than
those of Homo ergaster. The canine and the third molar are also reduced.
Furthermore, we have the morphology of a child who died at the age of
11, which revealed something surprising. Part of the frontal bone of this
child is preserved, mainly the right-hand half, which has a well-developed
brow ridge. There is no doubt that in an adult the ridge would be very
robust. Some of the transverse diameters of the front part of the skull have
been estimated; these suggest that the Gran Dolina child had a brain
larger than that of Homo ergaster (the brain grows very little after the
age of 11). In the three best-preserved skulls of Homo ergaster (ER 3833,
ER 3733, and WT 15000) the cranial capacity is respectively 804 cc,
850 cc, and 900 cc, while that of the Gran Dolina Child is at least 1,000 cc.
The face of the Gran Dolina Child is astonishingly modern. In Homo
habilis, Homo ergaster and, as far as we know, in Homo erectus, the skeleton
of the face is still fairly flat. Our face, however, is more sculpted, because
the nasal opening is forward of the rest of the face, and the bones of the
cheeks (the maxilla and zygomatic bone) are hollowed below the cheek-
bones, which thus project markedly. It is this combination of a primitive
frontal bone with a modern face that makes the Gran Dolina Child not
just one more fossil, but a very important specimen in our search for
information about our origins.
It was always thought that the modern face was a recent development in
human evolution, appearing with our species; suddenly we find that it
already existed 800,000 years ago. Where do we find fossils with a modern
face, of an age in between these two? We have the answer in Gran Dolina
itself, where fragments of the facial skeleton of adult individuals have also
been found, these showing a less sculpted form. We now know that as the
individual grew up, the face grew to become very large and robust, and
also more rounded as the maxillary sinuses expanded, eventually hiding
the features of the child in the adult’s face.
Hundreds of thousands of years later, our direct ancestors underwent
an enlargement of the brain which altered the structure of the neurocra-
nium, and a reduction in the masticatory apparatus, affecting the face, the
mandible, and the teeth. These are the two cranial traits which distinguish
us. The enlargement of the brain involved a fairly complete reorganization
of the neurocranium together with a marked change in its shape, but
the reduction in the masticatory apparatus was achieved in the simplest
manner possible: the facial skeleton stops developing at an earlier point,

Arsuaga / Chosen Species 1405115327-4-012 Final Proof page 189 4.6.2005 2:43pm

New Locations for Human Evolution

retaining a childish appearance. In other words, our adult face is like those
of the children of our ancestors.
So we see that the Gran Dolina fossils occupy an evolutionary position
intermediate between Homo ergaster and ourselves, the only modern
humans. Homo antecessor precedes our species but, as we shall see, it
also precedes the Neanderthals (Figure 12.2), another human species
different from ourselves which became extinct a few thousand years ago
(practically yesterday if we think in the vast terms of geological time, and
even in terms of the short duration of human evolution).
In principle it is assumed that the first humans reached the Iberian
peninsula by exclusively terrestrial means, coming from Asia and crossing
the whole of Europe. There is no reason to believe that the Straits of
Gibraltar closed at any point during the last 3 million years, although this
may well have occurred during a brief interval at the end of the Miocene,
between 6.5 and 5 million years ago (i.e., too early for humans, who did

Paleomagnetism Epochs
? H. sapiens
0 H. neanderthalensis

Middle H. rodhesiensis
H. heidelbergensis

H. erectus

Years (millions)

H. antecessor

−1 Jaramillo


? H. ergaster

Olduvai Pliocene


Figure 12.2 Evolutionary diagram proposed by the authors for the genus
Homo, from the initial settlement of Eurasia. The recently created species
Homo antecessor is included

Arsuaga / Chosen Species 1405115327-4-012 Final Proof page 190 4.6.2005 2:43pm

New Locations for Human Evolution

not yet exist, to cross over from Africa). As, moreover, the currents in the
Straits of Gibraltar are unfavorable for crossing, and the first humans are
not thought to have had seafaring skills, there are no solid arguments for a
western route, directly from Africa, for the colonization of Europe. We
shall see later that the oldest known human seafaring activity occurred a
few thousand years ago, and made possible the settlement of Australia and
Papua New Guinea. And those who achieved it were humans of our own
species (when sea level falls during the ice ages, Java and England can be
reached over land, but not Australia).
But if the European populations of Homo antecessor came from Asia,
and the Asian populations from Africa, where are the fossils of this species
outside of Europe? The answer is that they have not yet been found, partly
because there are no good African fossils of the same age, and the Asian
fossils which might be contemporary with Homo antecessor are those of
Homo erectus from the Far East. Three mandibles and a parietal bone were
found in North Africa, in the Tighenif (previously Ternifine) deposit in
Algeria; these are dated at 700,000–600,000 years old, later than the
Homo antecessor fossils. There are other mandibles of similar or slightly
more recent age in East Africa and Morocco. Unfortunately in Gran
Dolina we have only a fragment of an adolescent mandible to compare
with the African specimens.
We will therefore have to wait a little longer to get to know the African
relations of the Gran Dolina fossils. From that moment on, the European
branch of Homo antecessor, represented by the humans of Gran Dolina,
and the African branch, whose fossils have not yet been discovered,
followed different evolutionary histories.

Human Evolution in Europe during the Middle


Until recently, European fossil deposits were associated with one or other
of the cold or warm periods of the Alpine glaciations, identified on the
basis of the river terraces along the upper reaches of the Danube River.
However, as we have already noted in the chapter on climate, paleotem-
perature curves, established on the basis of the ratio of heavy and light
isotopes of oxygen in the calcareous shells of microfossils obtained from
deep-sea surveys, are now used. This record is much more consistent than
the continental record, and reflects changes of temperature at a global
level, since the ratio of oxygen isotopes is related to the size of the polar

Arsuaga / Chosen Species 1405115327-4-012 Final Proof page 191 4.6.2005 2:43pm

New Locations for Human Evolution

ice caps and the corresponding rises and falls in sea level. Paleotempera-
ture curves are divided into a series of isotopic stages, numbered from the
present backward. The present day is isotopic stage number 1. All the
odd-numbered stages are warm, and the even-numbered stages are cold.
The Middle Pleistocene is a division of the Pleistocene which runs from
the Matuyama/Brunhes change in magnetic polarity 780,000 years ago to
the beginning of isotopic stage 5. This warm stage began around 127,000
years ago, at a point when the climate was particularly warm (perhaps even
warmer than now). The Middle Pleistocene is a very important period in
human evolution because it was this period that saw the development of
the two human species we know best – Neanderthals and ourselves.
The oldest human fossils of the European Middle Pleistocene, apart
from those of Mauer and Boxgrove, are those of Arago (France) and
Ceprano (Italy), all dated at more than 415,000 years old (before isotopic
stage 11). The fossils found in the Arago cave include two mandibles, a
facial skeleton and a right parietal bone which belong to the same indi-
vidual, a coxal bone, and some other human remains. The incomplete
braincase from Ceprano was discovered in a ditch dug during construc-
tion of a freeway. Morphologically it has been described as very archaic,
close to Homo erectus. This is a very interesting specimen, but there are
still many questions relating to both its chronology and its phylogenetic
(evolutionary) position.
The Middle Pleistocene human fossils of intermediate age, situated
chronologically between isotopic stages 11 and 8 (from 415,000 years
ago to 245,000 years ago), form another group. These include the cranial
remains from Bilzingsleben, Steinheim, and Reilingen (Germany), Swan-
scombe (England), Petralona (Greece), Vértesszöllös (Hungary), and
Sima de los Huesos (another deposit in the Atapuerca Mountains in
Spain, which we describe in detail below).
The most recent European Middle Pleistocene fossils are those of warm
stage 7 (from 245,000 to 190,000 years ago) and cold stage 6 (between
190,000 and 127,000 years ago); this group includes, among others, the
remains from Ehringsdorf (Germany) and Biache-Saint-Vaast and La
Chaise-Abri Suard (France). In 1993 a complete skeleton covered with
calcareous concretion was found in the bottom of a pothole in the
Lamalunga cave (Altamura, Italy). This may be a late Middle Pleistocene
The teeth from Pinilla del Valle (Madrid, Spain) are also from the late
Middle Pleistocene. At other Spanish sites, the Bolomor molar (Valencia),
the Lezetxiki humerus (Guipúzcoa), the humerus and coxal bones from
Tossal de la Font de Vilafamés (Castellón), and the mandible and other

Arsuaga / Chosen Species 1405115327-4-012 Final Proof page 192 4.6.2005 2:43pm

New Locations for Human Evolution

human remains from Valdegoba (Burgos) may also be of this age. In

Bañolas (Gerona, Spain) a mandible with very unusual morphology (per-
haps related to the strange wear shown by the teeth) was found; this is
difficult to ascribe to any group. Some authors have attributed it to the
late Middle Pleistocene, although the travertine limestone in which it was
found is much more recent (around 45,000 years old). Finally, human
fossils have been found in the Cabezo Gordo deposit (Murcia, Spain);
some of these appear to be from the Middle Pleistocene.

The Pit of Bones

We have already noted that the discovery of a human fossil in a European

deposit of the Lower or Middle Pleistocene is something close to a
miracle. This is why these finds are so rare and so valuable. One of these
extremely rare and lucky chances (for us) occurred in Gran Dolina, and
the cause was an episode of cannibalism which took place about 800,000
years ago. But another site, also in the Atapuerca Mountains, holds the
largest deposit of human fossils ever found, and it owes its existence to
another extraordinary cause.
Not far from the trench of the abandoned railroad and from Gran
Dolina, there is an extensive system of underground caverns with two
entrances: Cueva Mayor (Mayor Cave) and Cueva del Silo (Silo Cave).
Following a rough path for 0.5 km from the mouth of Cueva Mayor, we
find a vertical shaft 14 m deep, which then continues for a further few
meters on a slope and ends in a small chamber. This cul-de-sac is known as
La Sima de los Huesos (The Pit of Bones), and indeed it contains a large
deposit of fossil bones, embedded in clay. The bones are all of carnivores
or humans: not a single herbivore fossil has been found, nor any stone
tools. Among the carnivores, the majority are bears of the species Ursus
deningeri, the ancestor of the enormous cave bear which lived thousands
of years later. There are probably over 200 of these bears accumulated in
La Sima. There are also remains of some lions, wolves, lynx, cats, foxes,
and Mustelidae of the marten and weasel type. In terms of human re-
mains, there are at least 32 individuals in the deposit. In both the carni-
vores and the humans all parts of the skeleton are represented, proving
that what accumulated in this pit were complete cadavers rather than
isolated bones.
The first human fossil in the Atapuerca Mountains was found here in
1976 by Trino Torres, an expert on fossil bones, and Edelweiss, a Burgos

Arsuaga / Chosen Species 1405115327-4-012 Final Proof page 193 4.6.2005 2:43pm

New Locations for Human Evolution

potholing group. This momentous discovery prompted Emiliano Aguirre,

who was at that time supervising Trino Torres’ doctoral thesis, to launch
an ambitious excavation project throughout the Atapuerca Mountains.
Unsupervised activities by visitors to La Sima before 1976 had altered the
upper sediments of part of the deposit, so that in the 1976 excavation, the
sampling carried out in 1983, and the campaigns of 1984 and 1988 work
was almost exclusively on sediment that had been disturbed. Removal of
this material was effectively completed by 1989, and since then the intact
sediment has been excavated in a series of annual campaigns.
The work to remove the disturbed sediment (in backpacks until 1987),
extract several tons of large limestone blocks, and adjust the infrastructure
of the cave, was a major undertaking. Moreover, the sediments disturbed
by visitors contained almost exclusively bear bones, with very few, in any
case very fragmented, human remains. But all of this work was rewarded
in 1992 with the discovery of three very complete skulls, including Skull
5, at this moment in time the best-preserved skull in the fossil record of
human evolution (Figure 12.3). Excavations in La Sima are still continu-
ing to produce a large quantity of very well-preserved human fossils.

Figure 12.3 Skull 5 from La Sima de los Huesos

Arsuaga / Chosen Species 1405115327-4-012 Final Proof page 194 4.6.2005 2:43pm

New Locations for Human Evolution

An extensive program to date the speleothems (mineral deposits) in La

Sima, as well as dating the human and bear remains directly, using the
uranium series and ESR methods, is underway in order to establish the
age of the human fossils. This work is being carried out by geochronolo-
gists James Bischoff and Christophe Falguères. The results obtained,
which are in keeping with the fauna, give an age of about 300,000 years
for the human fossils.
How did this deposit form? The cavities close to La Sima, within the
Cueva Mayor, also contain abundant remains of bears of the same species
as La Sima, which no doubt used these chambers to hibernate, and
sometimes died there from natural causes (sickness, age, or simply because
they had not accumulated enough fat in the autumn to last them through
the long winter). Some of the bears in the cave might have fallen acciden-
tally into La Sima, from which they would not have been able to escape.
The odd lion, wolf, or other creature, attracted by the odor of carrion,
might occasionally have jumped into this natural trap (there are other
deposits of this type at the base of potholes). At that time they would not
have had to make the long journey currently taken from the mouth of the
Cueva Mayor, since there was a much more direct entrance, which later
fell in.
However, the presence of at least 32 human cadavers in the deposit is
more difficult to explain than that of the bears. The absence of herbivores
and stone tools means that we can eliminate the possibility that humans
carried on their habitual activities here (even if there was at the time a less
rugged access route, apart from the pit of La Sima itself ). Nor does it seem
reasonable to attribute the accumulation of human cadavers to the action of
carnivores, again because of the absence of herbivores in the deposit. Two
alternatives remain. One is very vague: some kind of unidentified catas-
trophe. The other, which we prefer, is that other human beings, who could
have dropped the bodies of their dead companions into a pit in a hidden,
dark part of the cave, were responsible for the accumulation. This would be
the oldest known evidence of a funerary practice.
The paleodemography (distribution of ages of death) of the fossils from
La Sima has been studied by José Marı́a Bermúdez de Castro, as part of his
work in dental paleoanthropology. From this we know that most of the
individuals whose bodies accumulated in the deposit were adolescents
aged between 13 and 19, and young adults, less than 30 years old. This
age distribution is one more mystery among the many surrounding La
Sima. Why are there no children, or mature and old adults? Did all the
individuals die within a short period (months or years), or is the accumu-
lation the result of a practice which lasted for generations? Is it possible

Arsuaga / Chosen Species 1405115327-4-012 Final Proof page 195 4.6.2005 2:43pm

New Locations for Human Evolution

that everyone at that time died very young, before the age of 40? This
latter hypothesis is difficult to entertain, since we know that the period of
maturation in this fossil population was similar to our own, and we
therefore have to conclude that their potential life expectancy was roughly
the same as ours today. Even chimpanzees, whose period of maturation is
much shorter than that of humans, live beyond the age of 40 in the wild.
This is, in any case, just one of the problems currently being investigated;
it is also a matter of great interest in relation to finding out why so many
cadavers accumulated in La Sima de los Huesos.
The present authors, together with José Miguel Carretero, Ana Gracia,
and Carlos Lorenzo, have spent many years studying the human fossils of
La Sima; among their areas of interest is the difference between the sexes
in terms of body size. Many experts maintain that since sexual dimorph-
ism in body volume was much greater in the first hominids than it is now,
there must have been a more or less constant gradual reduction over the
course of human evolution. In this case, the Middle Pleistocene fossils,
including the Neanderthals of the Upper Pleistocene, should still show
greater sexual dimorphism than we do.
Two serious problems arise when we try to verify this hypothesis. One is
the proverbial scarcity of fossils, particularly of the postcranial skeleton,
which is used to estimate body size. Thus researchers are forced to
combine fossils from deposits in different places and eras in order to put
together a sample, which is in any case very small. The problem is not so
great in the case of the Neanderthals, for which there is a much larger
(though also dispersed) sample.
The second problem is methodological. In order to compare women
and men we first need to establish the sex of the fossils. The criterion of
size is generally the one used: the large bones are assumed to be those of
males, and the smaller bones to belong to females. But when the differ-
ence in size is not enormous, as it is in gorillas and orangutans, there is a
wide band of overlap of medium-sized and small male individuals and
large and medium-sized females, in which it is difficult to establish the sex
of the remains.
In La Sima de los Huesos we have a broad sample of individuals from a
single biological population. Moreover, our approach eliminates the
methodological problem, because what we are studying is the variation
in the sample, without assigning sex a priori to the fossils; we start from
the premise that the greater the difference between the sexes in the
population, the greater will be the variation observed in the sample.
In addition to measuring the bones of the postcranial skeleton, which
reflect differences in body weight, we also analyzed variation in brain size.

Arsuaga / Chosen Species 1405115327-4-012 Final Proof page 196 4.6.2005 2:43pm

New Locations for Human Evolution

In La Sima there are three skulls whose cranial capacity is known. One is
Skull 5, with a cranial capacity of 1,125 cc; the second is Skull 4, a
complete calvarium with a cranial capacity of 1,390 cc. The third is Skull
6, fairly complete, and corresponding to an adolescent aged about 14,
which has a cranial capacity of about 1,220 cc.
The result we have obtained is that the sexual dimorphism in body and
brain size in the population represented in La Sima (from about 300,000
years ago) was no greater than our own.
As La Sima de los Huesos is not a place in which humans would have
lived, we have no information on their behavior here. However, in one of
the Trinchera del Ferrocarril (Railroad Trench) deposits of the Atapuerca
Mountains, known as La Galerı́a (the Gallery), 13 archeological levels
have been excavated which give us some idea of human activity and
technology of the same period as La Sima and a little after. Two human
fossils, a jawbone fragment and a skull fragment, have also been found in
La Galerı́a.

Arsuaga / Chosen Species 1405115327-4-013 Final Proof page 197 4.6.2005 2:44pm


The Neanderthals

‘‘The Red Flower?’’ said Mowgli. ‘‘That grows outside their huts in the
twilight. I will get some.’’
Rudyard Kipling, The Jungle Book

The Way They Were

The discovery of the skull of a Neanderthal child aged two or three in

Engis (Belgium) in 1830 ought to have marked the beginning of paleo-
anthropology. But, like the Neanderthal skull excavated in the Forbes
quarry in Gibraltar in 1848, its true significance in human evolution was
not recognized at the time. Much more famous, and also controversial,
was the discovery in 1856 of a skeleton – which gives its name to the
entire group – in the Feldhofer cave near Düsseldorf, in the Neander
valley (Neander Thal in the old German spelling, Neander Tal in modern-
day spelling).
In 1859 Charles Darwin published On the Origin of Species, inching
open the door to the search for fossil antecedents of our species. A few
years later, in 1871, Darwin opened wide this door with his book The
Descent of Man, and Selection in Relation to Sex. Nevertheless, the status
of Neanderthals as members of an extinct human form distinct from our
own was not definitively recognized until the discovery of further fossils,
in particular those found in Spy (Belgium) in 1886, made it impossible
to continue considering them as atypical or pathological specimens of
modern humans.

Arsuaga / Chosen Species 1405115327-4-013 Final Proof page 198 4.6.2005 2:44pm

The Neanderthals

Many more hominid fossils, some of them of incomparably greater

antiquity, have been found since, but these have failed to usurp the place
occupied in the popular imagination by the Neanderthals, and they con-
tinue to be seen as the quintessential fossil humans. Nevertheless, the
Neanderthals were not the brutal, apelike beings, incapable of walking
erect, that many people suppose. They were strong and at the same time
skillful gatherers of plant products, hunters, and carrion eaters. They had a
wide variety of highly refined stone tools. They used fire systematically,
looked after their old and disabled, and buried their dead.
We know them well, almost as if they were still living, because in fact
Neanderthals are the human fossil type which has been studied in the
greatest detail, and of which we have the most remains. The Neanderthals
were not very tall, with the males averaging 170 cm in height, and the
females 160 cm. However, they had great physical strength. Christopher
Ruff, Erik Trinkaus, and Trenton Holliday have studied the body weight
of humans over the last 2 million years, and have concluded that the
Neanderthals were the humans with the greatest muscle mass – in other
words, the strongest (although preliminary studies, yet to be published,
indicate that the humans of La Sima de los Huesos were even stronger).
On average (for both sexes) Neanderthals weighed 70 kg, compared to
the average for our species (taken as a whole over the different living
populations and both sexes) of around 58 kg – 24 percent less.
The La Ferassie 1 fossil, from France, which is probably male, gives a
good example of the strength of the Neanderthals: it has an estimated
height of 171 cm, and a calculated weight of 85 kg. Moreover, John
Kappelman believes, in our view quite reasonably, that the weight esti-
mates made by Ruff and his colleagues are on the low side, because they
use formulas based on the relationship between body weight and bone
size in normal modern subjects. However, the development of the inser-
tions of the muscles in the bones indicate that the musculature was much
more developed in Neanderthals than in normal modern humans, and it
might have been more appropriate to make the comparison with elite
athletes, such as weightlifters, or javelin or discus throwers. As a result, it is
entirely credible that La Ferassie 1 weighed as much as 90 kg of pure
muscle, despite the individual being only 171 cm tall (Figure 13.1).
In the German deposit of Schöningen, a number of wooden spears around
400,000 years old have almost miraculously been preserved in the peat of an
ancient bog where humans hunted horses. The longest is 2 m long and was,
according to Hartmut Thieme, who discovered it, designed to be thrown.
We need only imagine a group of muscular Neanderthal hunters armed with
spears like these to see that they were hardly defenseless creatures.

Arsuaga / Chosen Species 1405115327-4-013 Final Proof page 199 4.6.2005 2:44pm

The Neanderthals

Figure 13.1 View of a Neanderthal with a wooden spear like those found in

Arsuaga / Chosen Species 1405115327-4-013 Final Proof page 200 4.6.2005 2:44pm

The Neanderthals

Many authors believe that this strapping physique corresponds to a

well-known law of biogeography known as Bergmann’s rule, according
to which the populations of a warm-blooded species living in cold regions
tend to have a larger body than those living in hot regions. In this way, the
body shape approaches that of a sphere, the form which has the smallest
surface area in relation to volume (the musk ox is a perfect example of an
animal from a polar environment). This relative reduction in surface area
reduces heat loss from the body through radiation.
In addition, the Neanderthals had relatively shorter forearms and lower
legs (below the knees). However, populations living in hot regions follow
the converse Allen’s rule, their long, thin arms and legs maximizing the
surface area to volume ratio (the dromedary is a good example of this
biotype). Within our species these laws are also observed, as can easily be
verified if we compare an Inuit from the Arctic Circle with a Tutsi or a
Tuareg, living in the hottest regions of Africa. In fact, the proportions of
the Turkana Boy (as the reader will remember, a Homo ergaster from 1.5
million years ago) would, according to his discoverers, be similar to those
of the human populations living in this region today (see Figure 7.5). The
first colonizers of Eurasia, who came from Africa, must also have been tall
and slim (a hypothesis which further discoveries of new fossils in Gran
Dolina will enable us to confirm); hundreds of thousands of years later the
Neanderthal physique reflected an adaptation to the rigors of a cold
climate, which enveloped much of the European continent during the
ice ages.
One notable characteristic of the Neanderthals is a very elongated and
flattened pubic bone (specifically, the upper, horizontal ramus of the pubic
bone). The elongation of the pubis is probably not a trait exclusive to the
Neanderthals, since it is also found in australopithecines, as we have seen.
Before any complete Neanderthal pelvises had been found, and only
fragments of pubic bones and other broken pieces were available, some
authors believed that, given the extraordinary length of the pubis, the
birth canal would also be very large. There were even some who suggested
that pregnancy would have been a month or so longer than in our species,
so that Neanderthal babies came into the world more fully developed.
However, the discovery of a very complete Neanderthal pelvis in the
Kebara deposit in Israel showed that although the pelvic structure was
different from ours, the birth canal was not substantially larger – perhaps a
little larger, but since the adult brain was also a little larger than ours, the
stage of development of the newborn would have been comparable with
that of our babies. For new information on this question we will have to
await publication of the results of studies currently under way on the

Arsuaga / Chosen Species 1405115327-4-013 Final Proof page 201 4.6.2005 2:44pm

The Neanderthals

complete pelvis recently discovered in the Sima de los Huesos deposit in

Atapuerca, Spain.
It may have come as a surprise to the reader that Neanderthals’ brains
were not smaller, but larger than ours: the average capacity was 1,500 cc,
whereas the modern average, calculated for all populations, is around 1,350
cc. However, given that because of their great muscle mass the body weight
of the Neanderthals was higher, it is likely that the index of encephalization
(the only way we have in paleontology to express, in terms of measurement,
something like intelligence) was slightly lower than our own.
What is certain is that Neanderthals and ourselves are the two most
highly encephalized human forms in history. However, this increase in
brain size occurred independently in the two cases. While the modern
neurocranium is high, with a high forehead, that of the Neanderthals was
very elongated from front to back (Figure 13.2). Because of this anterio-
posterior elongation of the brain cavity, the occipital bone extended
backward, forming the characteristic bump known as the ‘‘occipital
bun’’ of Neanderthal skulls, which is clearly visible from the side. The
paleoanthropologist Giorgio Manzi has shown how the increase in brain
size in the Neanderthals did not involve a substantial change in the archaic
structure of the skull, which remained low, with a flat forehead; it was
simply modified to contain a brain which was sometimes enormous.
Manzi has said that in Neanderthals the enlargement of the brain caused
a conflict similar to that which would be generated if a small saloon car
had to be adapted to take a Formula 1 engine.
Viewed from the back, there is an important modification in the neu-
rocranium of the Neanderthals. In Homo ergaster and Homo erectus the
skull was widest at the base, and the two sides converged toward the top of
the cranial vault, like a house with the walls leaning inward. In Neander-
thals and modern humans the widest part of the neurocranium is in the
middle, above the parietal bones; the skull is narrower at the base. How-
ever, while our skull, viewed from the back, looks like a house with the
walls leaning outward from the base toward the roof (i.e., inclined out-
ward), that of the Neanderthals is rounded.
Neanderthals show other neurocranial traits which are exclusive to
them, some of them not very noticeable but extremely useful for tracing
the origins of these humans. For example, above a not very marked
occipital ridge, sunken in the center, Neanderthals have a depression
technically known as the suprainiac fossa. To cite just one other character,
in Neanderthals the mastoid process (the bony projection of the temporal
bone, where part of the sternocleidomastoid muscle originates) projects
hardly at all beyond the base of the skull (Figure 13.2).

Arsuaga / Chosen Species 1405115327-4-013 Final Proof page 202 4.6.2005 2:44pm

The Neanderthals


Occipital ridge

Mastoid process

Retromolar space


Figure 13.2 Neanderthal skull (top) and modern human skull (below)

Arsuaga / Chosen Species 1405115327-4-013 Final Proof page 203 4.6.2005 2:44pm

The Neanderthals

Neanderthals have an easily recognizable brow ridge: it forms a regular

curve above the eye sockets like an arc of a circle, it is rounded in cross-
section and it continues between the eyes. Moreover, in general it is
completely hollowed out inside by highly developed frontal sinuses.
Finally, the Neanderthal face is unique among the hominids (Figure
13.3). In other fossils, and in ourselves, the bones below the eye sockets
and either side of the nasal opening (the maxilla and the zygomatic bone)
form a bony surface which faces forward. We have already seen how in our
species the lower part of this surface is hollowed, allowing the cheekbones
to project above. In Neanderthals, in contrast, this surface faces diag-
onally, giving them a wedge-shaped face. It is as if the nasal opening had
moved forward, stretching the sides of the face (this typical morphology
of the Neanderthal face is technically known as medio-facial prognath-
ism). In the foremost part of this pointed face there is a very wide nasal
Thus Neanderthals have a large nasal cavity (with its roof formed by
nasal bones which are almost horizontal, as a result of the forward shift of
the nasal opening), which some authors have interpreted as an adaptation
to a glacial, cold, dry climate. The air would be warmed and humidified in
this cavity before passing into the lungs. Moreover, the highly developed
frontal sinuses (above the nasal opening and over the eye sockets) and
maxillary sinuses (either side of the nasal cavity) would contribute to form
an air chamber which would insulate the brain, an organ which is very
sensitive to changes in temperature. An enlarged face, like a great hollow
mask, would thus be interposed between the brain and the cold outside.
There are authors, however, who interpret the facial morphology of the
Neanderthals differently, in biomechanical terms (although the two ex-
planations are in fact compatible). The front teeth of the Neanderthals
show very rapid wear, indicating intense use, as if the mouth was often
used as a ‘‘third hand’’ to hold and pull on objects. The pointed shape of
the face, it is suggested, would serve to divert the force generated in the
bone by this activity toward the sides.
Another characteristic which is almost exclusively Neanderthal is that in
the jawbone the teeth are shifted forward in relation to the bone, to the
extent that there would be room for a fourth molar. This empty space,
known as the retromolar space, means that when seen from the side the
Neanderthal jawbone has a hollow between the last molar and the front
edge of the mandibular ramus (the branch of the jawbone which runs
upward toward the joint with the base of the skull).
In the living Neanderthal, this face would have been distinguished by
very wide, prominent nostrils, the absence of cheekbones, a sloping

Arsuaga / Chosen Species 1405115327-4-013 Final Proof page 204 4.6.2005 2:44pm

The Neanderthals

Figure 13.3 Reconstruction of the head of a Neanderthal

Arsuaga / Chosen Species 1405115327-4-013 Final Proof page 205 4.6.2005 2:44pm

The Neanderthals

forehead, strongly overhanging brows, and a jaw with no chin (Figure

13.3); a well-developed chin is a trait exclusive to ourselves (Figure 13.2).
Finally, the Neanderthals’ bones were thick, although this is a charac-
teristic found in all humans since Homo ergaster, except in our species,
where the walls of the bones have become markedly thinner. But while in
Homo ergaster the bones of the head were the thickest, in Neanderthals
the thickest bones were those of the body; the medullary cavities, which
are found inside the long bones like the femur, humerus, and tibia, are
very narrow as a result of the thickening of the bone walls. The reason for
the production of so much bone is not actually known, but it is certain
that the Neanderthals needed a great deal of calcium. This is not found in
meat: in fact an excess of animal proteins acidifies the blood, requiring
neutralization by the release of calcium from the skeleton and thus de-
pleting the mineral levels in the bones (carnivores have bones with thin
walls, while in herbivores they are thicker). One source of calcium is milk
and dairy products, which Neanderthals certainly did not eat. Their only
source of calcium, therefore, was plants; Neanderthals no doubt had to
consume these in large quantities.

Life and Death among the Neanderthals

The Neanderthals’ stone tool industry is known as Mousterian, and is

classified within Mode 3 or the Middle Paleolithic (Figure 13.4). The
characteristic of this technical mode is that the cores are worked by giving
them a specific shape (similar to a turtle shell), in order to obtain flakes
from them, which will then be retouched to achieve the final form. This
sequence of operations is known as the Levallois technique, and various
tools are obtained from each core, thus making better use of raw material
and effort. Moreover, it is clear that it implies a distinct capacity for
abstraction, since the stone is not worked directly to produce the tool;
instead an intermediate step (the Levallois core) is introduced.
The Mousterian extended throughout Europe, the Near East, and
North Africa, around the Mediterranean, while other Mode 3 or Middle
Paleolithic industries are found in the rest of Africa (where they are
grouped together under the term ‘‘Middle Stone Age’’). Mode 3 origin-
ated between 300,000 and 200,000 years ago, depending on the region,
appearing first in sub-Saharan Africa and later in Europe. Once more we
come up against the problem of how the technique arrived in Europe. Was
it brought by people who came from Africa? And if so, what was their

Arsuaga / Chosen Species 1405115327-4-013 Final Proof page 206 4.6.2005 2:44pm

The Neanderthals

Figure 13.4 Some representative Mode 3 tools. Top: Levallois core and point
extracted from it; below left : retouched scraper of the la Quina type; below right :
bifacial point (Middle Stone Age)

relationship with the local populations, those responsible for the Acheu-
lian (Mode 2) industry? Or was it only the new technique that traveled,
without population movements? We shall return to this problem later.
Two aspects of Neanderthal behavior which bear striking similarity to
our own are the use of fire and the practice of burying the dead. There are
animals which, albeit in very simple ways, select and even modify natural
objects to use them as tools. However, no animal species other than ours
knows the technology of fire, nor does any bury its dead or conduct any
ceremony for them. The practice of burial is thus a ‘‘humanizing’’ trait. It
is difficult now to imagine human life in the wild without the use of fire.
In fact, what is surprising is that there have been humans who did not have

Arsuaga / Chosen Species 1405115327-4-013 Final Proof page 207 4.6.2005 2:44pm

The Neanderthals

fire, at least in Europe and parts of Asia (in regions far from the equator).
In many senses fire makes us human, and the absence of fire places us on
the same level as animals.
It is not known for certain when the capacity to make and control fire
first appeared among humans. It is possible, although not certain, that it
was used in the Middle Pleistocene in places like Zhoukoudian in China,
Terra Amata in France, Vértesszöllös in Hungary, La Cotte de St. Brelade
(on the island of Jersey, Great Britain), and Bilzingsleben in Germany – all
deposits which are 200,000 years old or more. However, the generalized
and systematic use of the technology of fire, with all it implies in terms of
protection, heat, light, and so on, emerged somewhat less than 200,000
years ago. From this time on there are well-structured fireplaces in the
deposits, which leave no room for doubt that fire had been mastered. At
this point the humans living in Europe, those who made the fires, were
Neanderthals, now masters of the ‘‘Red Flower.’’
It has always seemed evident that the Neanderthals buried their dead,
and many of the Neanderthal skeletons excavated in cave deposits have
historically been considered the result of this burial practice (for example,
eight Neanderthals were found entombed in the French deposit of La
Ferassie). It is less certain whether the burials were accompanied by a
ritual, a ceremony with some symbolic significance, although in some
cases it has been suggested that there is evidence of this. For example,
the earth covering the skeleton of one of the Neanderthals from Shanidar
(Iraq) contained grains of pollen from flowers which were thought to have
been placed on the body; several pairs of mountain-goat horns found
around the skeleton of a nine-year-old child in Teshik Tash cave (Uzbeki-
stan) are thought to have been placed there deliberately; on the skeleton
of a two-year-old child in Dederiyeh (Syria) a triangular flint tool was
found at the level of the heart and a limestone ‘‘tombstone’’ next to
the head; a deer jawbone was found on the pelvis of a ten-month-old
child at Amud (Israel); the Le Moustier adolescent, in France, was
thought to have been sprinkled with ochre and buried in a flexed position,
accompanied by offerings.
However, in their book In Search of the Neanderthals, paleoanthropolo-
gist Chris Stringer and archeologist Clive Gamble very reasonably call into
question this evidence of ritual, which of course is open to other, more
prosaic interpretations. A casual association of human bones with animal
remains or tools, which produces the appearance of something inten-
tional, is always possible in a deposit where there is an abundance of
animals and stone tools (and pollen is brought in by the wind). Moreover,
some of the excavations, such as that of Le Moustier, are old, and there are

Arsuaga / Chosen Species 1405115327-4-013 Final Proof page 208 4.6.2005 2:44pm

The Neanderthals

reasons to question how strictly they were conducted (each generation of

scientists criticizes the methodology of previous generations, although
not always with justification). Stringer and Gamble extend their skepti-
cism even to the suggestion that Neanderthals buried their dead.
We believe that they did do so, because the deaths of Neanderthals
whose skeletons have been found in caves cannot all be due to natural
causes. Furthermore, there is evidence that the Atapuerca humans were
already conducting funerary practices (though not burials exactly)
300,000 years ago, in accumulating the bodies of their dead in La Sima
de los Huesos. It is true that this is a unique case, but it is also possible that
burials or other funerary practices were conducted at this time outside of
caves, leaving us no fossil record or knowledge of them.

The Beginning and End of the Neanderthals

The Neanderthals properly speaking, i.e., with all or the majority of their
typical characteristics, existed in Europe some 230,000 years ago. The
fossils from Ehringsdorf, Biache-Saint-Vaast, la Chaise-Abri Suard, and
others from the end of the Middle Pleistocene can already be considered
The key to understanding the evolutionary position of the group of
Middle Pleistocene fossils of intermediate age (between 415,000 and
245,000 years old), such as those of Steinheim, Swanscombe, Reilingen,
Vértesszöllös, Petralona, and La Sima de los Huesos, is in the broad
sample from the Spanish deposit. Although in general their morphology
is primitive, the fossils from La Sima de los Huesos show incipient Nean-
derthal traits in the occipital bone and the face, and other more clearly
Neanderthal features in the mandible, for example the presence of a
retromolar space (see Figure 12.3). The postcranial skeleton also bears
traits in common with the Neanderthals, such as the morphology of the
pubis and humerus, among others. But there are so few remains of these
bones in the fossil record outside of La Sima de los Huesos that at present
it is difficult to know at what point in human evolution these traits
appeared (it may even have been before the colonization of Europe).
The Steinheim skull may possibly have been similar to those of La Sima
de los Huesos, though it is difficult to tell because it is incomplete and
badly deformed. More fragmented fossils, such as those from Swan-
scombe, show a more typically Neanderthal occipital morphology (with
a ridge that is depressed in the center and a broad suprainiac fossa). The

Arsuaga / Chosen Species 1405115327-4-013 Final Proof page 209 4.6.2005 2:44pm

The Neanderthals

Petralona skull has a more Neanderthal face than those from La Sima de
los Huesos, but its occipital bone is less Neanderthal. Taken together, this
group of fossils indicate that the Neanderthals evolved in Europe over
hundreds of thousands of years in geographical and genetic isolation. In
this sense the Neanderthals are the original Europeans, a native, local
human species – what is known in biology as an endemic species.
The Neanderthals’ roots extend far back into the past, but how far back
do they go? Some authors, such as Chris Stringer, group the Mauer
mandible with the more modern fossils of Arago, Bilzingsleben, and
Petralona together to form a species known as Homo heidelbergensis,
with the Mauer mandible as the type specimen.
African specimens dated at between 600,000 and 250,000 years old,
such as the skulls of Bodo (Ethiopia), Ndutu and Eyasi (Tanzania), Salé
(Morocco), Elandsfontein (South Africa), and Broken Hill (Zambia), are
added to the European fossils of this species.
It is also suggested that Homo heidelbergensis is represented in Asia by
the Dali skull and the Jinnishuan skeleton, both from China, and dated at
between 200,000 and 300,000 years old. Precise dating of these fossils is
important, because they might be contemporaneous with the last Homo
erectus fossils in China, Skull 5 from Zhoukoudian, and an incomplete
skullcap from Hexian cave.
According to this theory, the place of Homo heidelbergensis in human
evolution would be that of the last common ancestor of Neanderthals and
modern humans (Figure 13.5). The cranial capacity of the species ranges
from 1,000 to 1,400 cc, the lower value in the range approximating to the
average for Homo erectus, and the upper to the average for Neanderthals
and modern humans.
In a recent article Robert Foley and Marta Lahr elaborate an evolution-
ary model which they call the Mode 3 hypothesis. According to them, the
species which invented Mode 3, between 300,000 and 250,000 years ago,
was a species which evolved in Africa from Homo heidelbergensis. They
suggest that this new species (whose Latin name we shall omit, so as not to
add further to the terminological confusion) had a larger brain and was
more intelligent than previous species, as demonstrated by their advanced
(Mode 3) technology. Some members of the Mode 3 Species migrated to
Europe around 250,000 years ago, in time replacing the European popu-
lation of Homo heidelbergensis and evolving into the Neanderthals. In
Africa the Mode 3 Species evolved to create our species, Homo sapiens.
However, in our view all these authors are mistaken. Our analysis of
morphological traits leads us to conclude that the European populations
such as that of Mauer, living around 500,000 years ago, were already on

Arsuaga / Chosen Species 1405115327-4-013 Final Proof page 210 4.6.2005 2:44pm

The Neanderthals

Paleomagnetism Epochs
? H. sapiens
0 H. neanderthalensis

H. heidelbergensis

H. erectus
Years (millions)

−1 Jaramillo


? H. ergaster

Olduvai Pliocene


Figure 13.5 Alternative evolutionary diagram to that of the authors, in

which Homo heidelbergensis appears as common ancestor of the Neanderthals
and modern humans

the evolutionary line of the Neanderthals. The rules for creation of species
state that the name of the species should be related to its type specimen;
as the type specimen for Homo heidelbergensis is the Mauer mandible,
this species should actually be exclusively European and the ancestor of
Neanderthals. The last common ancestor of Neanderthals and modern
humans is Homo antecessor, created on the basis of the Gran Dolina
fossils, which are more than 780,000 years old (see Figure 12.2). As we
shall see later, there are data from molecular biology which support
our position.
The species Homo heidelbergensis, as we understand it, would cover
fossils from the Mauer mandible to the fossils of La Sima de los Huesos
and all those in which primitive traits predominate, although they may
show some incipient characters indicating that they are ancestors of the
Neanderthals. The fossils of the end of the Middle Pleistocene (from
about 230,000 years ago onward), on the other hand, can to all intents
and purposes be considered true and complete Neanderthals.

Arsuaga / Chosen Species 1405115327-4-013 Final Proof page 211 4.6.2005 2:44pm

The Neanderthals

The two skulls from Saccopastore (on the outskirts of Rome) and the
set of fossils from Krapina (Croatia) are from the beginning of the Upper
Pleistocene (the latter have been reliably radiometrically dated, but the
date for the Saccopastore fossils is less certain).
From this time on Neanderthals appear in abundance in European
deposits (relative, of course, to the general scarcity of the fossil record),
and they are also found in Central Asia and the Middle East, regions to
which they migrated from Europe. Fossils as emblematic as those of Le
Moustier (which gives its name to the Mousterian), Guattari 1 (Monte
Circeo), and La Chapelle-aux-Saints lived in Europe less than 60,000
years ago. These Neanderthals show some new characteristics with respect
to earlier specimens, and are often called ‘‘classic Neanderthals.’’ The
description of a Neanderthal that we gave earlier corresponds most par-
ticularly to these classic forms. As we shall see, there were still Neander-
thals in Europe 30,000 years ago, perhaps even more recently, before their
trail disappeared forever.
The Spanish record of the Upper Pleistocene is rich in Neanderthals
and includes fossils from the deposits of Agut (Barcelona), Axlor (Viz-
caya), Cova Negra (Valencia), Gibraltar (Devil’s Tower and Forbes
Quarry), Gabasa (Huesca), La Carihuela (Granada), Los Casares (Guada-
lajara), Mollet I (Gerona), and Zafarraya (Málaga). There is good reason
to hope that many more Neanderthals will be discovered in the Iberian
peninsula over the next few years.

Arsuaga / Chosen Species 1405115327-4-014 Final Proof page 212 4.6.2005 2:44pm


The Origins of Modern

Humanity: The Fossil

Resolve me strangers, whence, and what you are;

Your business here; and bring you peace or war?
Virgil, Aeneid

Neanderthals and Modern Humans

In the European deposits which contain unbroken archeological se-

quences, the Mousterian (Mode 3) industry is abruptly replaced by the
Upper Paleolithic Aurignacian, or Mode 4, industry (Figure 14.1). Tech-
nically, Mode 4 is characterized by the preparation of elongated cores to
obtain fine flakes and parallel edges, at least twice as long as they are wide.
These flakes were then retouched and made into a wide variety of tools,
including burins (beveled tools used to work bone, horn, and marble) and
end-scrapers (flakes with one end retouched, used to prepare hides). This
technique allowed for maximum use of the raw material, obtaining more
blade length (if we add together the length of the blades of all the tools
made from the original raw material) than any other method. Mode 4 is
also distinguished by the use of bone, ivory, and horn as raw materials for
tools and items of personal adornment. Art also appears in association
with Mode 4 industries, in the form of portable figures of animals and
people, and cave paintings and engravings (although these artistic expres-
sions are not found with the first Mode 4 industries, appearing only some
thousands of years later).

Arsuaga / Chosen Species 1405115327-4-014 Final Proof page 213 4.6.2005 2:44pm

The Fossil Evidence

Figure 14.1 Representative Mode 4 tools. Top: end-scraper; left : bone spear;
right : double-beveled burin

However, in some deposits on the Atlantic coast of the Iberian penin-

sula and in western and central France, intercalated between the last
Mousterian and the first Aurignacian levels, we find intermediate levels
containing an industry with characteristics common to both technical
modes (3 and 4). This industry is known as the Chatelperronian; the
Szeletian industry of central Europe and the Uluzzian industry of Italy
may be equivalents. The Chatelperronian is an evolution of the Mouster-
ian industry which includes elements of Mode 4, such as long flakes, and
tools made from bone and ivory.

Arsuaga / Chosen Species 1405115327-4-014 Final Proof page 214 4.6.2005 2:44pm

The Fossil Evidence

The Aurignacian spread quickly through Europe, about 40,000 years

ago or a little earlier. We have two examples which have been studied in
depth where the Mousterian is abruptly replaced by the Aurignacian, in
the deposits of L’Arbreda and El Abric Romanı́ in Catalonia, Spain; here
radiometric data obtained by James Bischoff indicate that the change
happened around 40,000 years ago. On the Atlantic coast the oldest
Aurignacian industry, found in the El Castillo cave, has been dated by
Bischoff at 39,000 years old. Who made the Chatelperronian and the
Aurignacian tools?
The human fossils found so far with Mousterian (Mode 3) tools in
Europe are always Neanderthals, while those found with Mode 4 indus-
tries are always modern humans. Fossils associated with the Chatelperro-
nian industry include those of the Grotte de Renne at Arcy-sur-Cure (near
Auxerre, France), which consist of isolated teeth and bone fragments, and
the partially preserved skeleton from Saint-Césaire (Charente-Maritime,
There is no doubt that the Saint-Césaire fossil is of a classic Neanderthal
individual who lived around 36,000 years ago. The Chatelperronian level
associated with the human remains in the Grotte de Renne at Arcy-sur-
Cure is dated at 34,000 years old, and it contains bone and ivory tools,
together with items of personal decoration such as perforated or grooved
animal teeth and ivory beads or pendants (all decorative pieces typically
associated with Mode 4). The nature of the human fossils in the Grotte de
Renne was not clear because they were so incomplete. However, Fred
Spoor, Jean-Jacques Hublin, and their colleagues have demonstrated that
the morphology of the bony labyrinth (inner ear) of Neanderthals was
different to our own. This led to the identification of the temporal bone of
a child aged approximately twelve months as Neanderthal (the bony
labyrinth is fully formed before birth – fortunately, since it is impossible
to establish from the external morphology whether a temporal bone from
such a young individual is Neanderthal or not).
Thus, in the only two deposits which contain human fossils in association
with Chatelperronian industry, the fossils are of Neanderthals. On the
other hand, the very recent dates of the Chatelperronian, later than the
first Aurignacian industries, rule out the possibility that Neanderthals
evolved into modern humans in western Europe, if we assume that modern
humans were responsible for the first European Mode 4 industries.
The cave paintings, the most spectacular artistic expressions of prehis-
tory (and perhaps of the entire history of art), also begin early: the
marvelous images of animals recently discovered in the Grotte Chauvet
(in Ardèche, France) have been radiocarbon dated at about 30,000 years

Arsuaga / Chosen Species 1405115327-4-014 Final Proof page 215 4.6.2005 2:44pm

The Fossil Evidence

old, and no one would venture to attribute them to Neanderthals. Other

cave paintings, also recently discovered, in the Grotte Cosquer (Bouches-
du-Rhône, France), are about 27,000 years old. The oldest Upper Paleo-
lithic figurines, the stone figure of a woman from Galgenberg (Austria)
and the small ivory sculpture of a horse from Vogelherd (Germany), are
perhaps 32,000 years old.
However, unfortunately we do not as yet have any human fossils
associated with the first Aurignacian industries in Europe. The modern
skeletons from the Cro-Magnon deposit (France) are assigned to the end
of the Aurignacian, between 30,000 and 28,000 years ago. An interesting
fossil in this respect is the Hahnöfersand frontal bone (Germany) which,
although it has no context, has been radiocarbon dated at 36,000 years
old. As we understand it, the morphology of the brow ridge indicates that
it should be classified as a modern human; it is a shame that its age is so
problematic. In central Europe the modern skeletal series from Mladec (in
Moravia, Czech Republic) is dated at about 32,000–30,000 years old, and
is associated with an Aurignacian industry.
The last Neanderthals (like the Saint-Césaire specimen) show no signs
of evolution toward modern humans; in fact they are absolutely ‘‘classic’’
Neanderthals. Similarly, the Mladec remains cannot be considered transi-
tional forms: they are entirely modern, albeit robust, humans. This means
that the possibility of modern humans in Europe originating through
evolution from the local Neanderthals can also be ruled out categorically
from the paleoanthropological point of view. We have to conclude that
modern humans in Europe are immigrants who came from elsewhere.
Study of the body proportions of the first modern humans in the Euro-
pean record gives us a clue to their origins: their gracile biotype resembles
that of people from equatorial regions, less stocky, taller, and with longer
arms and legs than the Neanderthals.
The last Mousterian industries, which are found in Portugal, Spain, and
Italy, and are dated at exactly 30,000 years old, are contemporary with or
even later than the modern skeletons of Mladec. Archeologists Gerardo
Vega and Valentı́n Villaverde have pointed out that these Mousterian
industries survived in the Iberian peninsula after they had disappeared
from other regions of Europe. Confirming these observations, archeolo-
gist Cecilio Barroso and paleoanthropologist Jean-Jacques Hublin have
found some classically Neanderthal human fossils in the Zafarraya deposit
(Málaga, Spain); these include a mandible and a femur dated at 30,000
years old and associated with a Mousterian industry.
It is thus certain that the last Neanderthals were still living in southern
Europe when modern humans were already painting rhinoceros, lions,

Arsuaga / Chosen Species 1405115327-4-014 Final Proof page 216 4.6.2005 2:44pm

The Fossil Evidence

and bison on the walls of the Chauvet cave, and 10,000 years after the first
modern humans settled in the Iberian peninsula. However, this is not to
give the impression that modern humans advanced from east to west like a
steamroller, crushing all the Neanderthals in their path. The two species
must have shared the map of Europe for thousands of years, forming a
mosaic of intermingled populations. As we have seen, about 40,000 years
ago there were already modern humans in the regions of Catalonia and
Cantabria in Spain, and thousands of years later Neanderthals still sur-
vived well to the north of the Pyrenees. We may imagine that the pockets
of Neanderthal populations gradually diminished until the last of them
This long coexistence of Neanderthals and modern humans (popularly
known as Cro-Magnon Man) is one of the chapters of prehistory which has
most excited popular attention, giving rise to tales such as La Guerre du Feu
(Quest for Fire, 1911) by J. H. Rosny-Aı̂né, made into a film by Jean-
Jacques Annaud in 1981, and Jean M. Auel’s saga The Clan of the Cave
Bear. One particularly intriguing aspect of this coexistence between Nean-
derthals and modern humans is the origin of the Chatelperronian and other
similar technologies. Had the Neanderthals developed the use of bone,
horn, and ivory to make tools and decorations, and the production of long,
thin stone blades, of their own accord? Did they do it independently in a
number of regions of Europe? Or did they copy it from modern humans by
watching them work, or examining abandoned objects? Might some of the
elements found in the Grotte de Renne at Arcy-sur-Cure, such as the ivory
pendants, come from an exchange between the two types of humans? As yet
we have no solution to these puzzles.

Two Intelligent Human Species

From the paleontological point of view, Neanderthals are a different species

(Homo neanderthalensis) from modern humans (Homo sapiens). This means
that Neanderthals are the result of a long process of evolution independent
of our own, originating from a common ancestor. As a result of this
separate, divergent evolution, the difference between Neanderthals and
modern humans is much greater than between the various modern popu-
lations (Inuit, Aboriginal Australians, Zulus, and Basques, for example).
When defining a new living species, the genetic criterion is usually used:
members of a new species cannot mate with those of another species and
have fertile offspring which could in their turn reproduce with individuals

Arsuaga / Chosen Species 1405115327-4-014 Final Proof page 217 4.6.2005 2:44pm

The Fossil Evidence

from the population of the father or the mother, or with one another. The
new species must be genetically isolated from others. In the case of the
Neanderthals, we know of no fossil hybrids of them and our ancestors;
moreover, modern Europeans do not carry genes inherited from the
Neanderthals. However, the fact that there was little exchange of genes
does not mean that it was impossible, and this is the condition for defining
a species on the basis of genetic criteria.
In paleontology, the evolutionary concept of the species proposed by
George Gaylord Simpson (1902–84) makes more sense. According to this
concept, a species is a continuum of populations which succeed one
another in time and follow their own evolutionary trajectory, independent
of other species, and continue over a considerable period: what counts is
that there is genetic continuity between generations and that isolation is
maintained. According to this criterion, Neanderthals would be an ‘‘evo-
lutionary species.’’ Something similar occurred with the two species of
chimpanzee, which are separated by the Congo River: common chimpan-
zees live to the north, and pygmy chimpanzees or bonobos to the south.
The different species of gibbon also developed through geographical
However, according to Simpson’s criterion the species Homo heidelber-
gensis, as we understand it, would not exist, since fossils like those of La
Sima de los Huesos are the ancestors of the Neanderthals and therefore
belong to the same ‘‘evolutionary species’’: they are simply very primitive
Neanderthals. While we recognize the validity of this argument, we be-
lieve that given the morphological difference between the two types of
fossils, the species Homo heidelbergensis should be retained for practical
reasons, as is normally the case in paleontology (and particularly if we look
at things from the common-sense point of view, which dictates that we
should give different names to things which are distinct from one an-
On the other hand, if the Neanderthals were not human, no one would
dispute that they deserve their own species. However, many researchers
find it difficult to accept that human beings who buried their dead and
used fire were of a different species. What is more, European Neanderthals
even came to manufacture tools similar to those of modern humans (the
Chatelperronian industry). Lastly, we shall see below that the industry of
the first modern humans (in the Levant) was the same as that of the
Neanderthals (Mousterian).
However, all of these coincidences only mean that Neanderthals were
intelligent. The fact that we are the only intelligent human species cur-
rently in existence does not mean that it has always been so. We are also

Arsuaga / Chosen Species 1405115327-4-014 Final Proof page 218 4.6.2005 2:44pm

The Fossil Evidence

now the only bipedal primate species, and we have seen that in the past
there were several species of bipedal hominids living at the same time. The
Neanderthals represent another intelligent human species, among other
reasons because the common ancestors of Neanderthals and modern
humans (Homo antecessor) were also intelligent. There is nothing to rule
out the possibility that the different intelligent human species exchanged
information, produced the same type of tools, and shared the technology
of fire; two intelligent species can exchange information without exchan-
ging genes.
To put it another way, if an alien spaceship landed on our planet the day
after tomorrow, the beings which came out of it would no doubt be
intelligent, and we would find a way to communicate with them. Never-
theless, because they evolved in a different place, the aliens would be of a
different biological species from ours. Something similar happened, albeit
with fatal consequences for the Neanderthals, when our ancestors arrived
on the European continent for the first time.
In any case, assuming that technical change came with biological
change, as some researchers advocate, would imply four separate succes-
sive colonizations of Europe: first the manufacturers of Mode 1 (repre-
sented by the fossils of Gran Dolina) would arrive, then those who made
Mode 2 (Homo heidelbergensis), then those of Mode 3 (the ancestors of
the Neanderthals), and finally those of Mode 4 (modern humans). We
believe, on the contrary, that cultural diffusion between human popula-
tions or species was more frequent than replacement of one by another,
and therefore that Europe was settled only twice: first by Homo antecessor
about 800,000 years ago, and then by our ancestors around 45,000 years
There are researchers who believe they see a substantial difference, a
chasm, between the Neanderthal mind and that of modern humans. Thus
our superior cognitive capacity is manifested by, among other things, the
fact that we are the only ones capable of devising aesthetic and symbolic
concepts. We have already seen that there is debate as to whether the
Neanderthals buried their dead with ceremony. For us the mere fact that
they buried them already implies a ritual, and therefore a capacity for
On the other hand, we have to acknowledge that the explosion of art
occurred in the Upper Paleolithic. Before this there is some very doubtful
evidence, such as the series of cuts on bones found in the Bilzingsleben
deposit in Germany, over 350,000 years old, or the supposed sculpture of
a woman from Berekhat Ram in Israel, which is over 230,000 years old.
However, although the ancestors of the Neanderthals did not have art,

Arsuaga / Chosen Species 1405115327-4-014 Final Proof page 219 4.6.2005 2:44pm

The Fossil Evidence

nor did our ancestors at that time. The great examples of artistic expres-
sion like the paintings and engravings on walls and stone panels, and the
sculptures of animals and people, are found not in the early part of the
Upper Paleolithic, between 50,000 and 45,000 years ago, but only from
just before 30,000 years ago. It is possible that the Neanderthals did not
come to produce art simply because they became extinct before artistic
activity became widespread. However, we have already noted that 34,000
years ago the Neanderthals of the Grotte de Renne had items of decor-
ation similar to those of their modern human contemporaries, indicating
that they had an aesthetic sense.
But . . . if modern humans did not evolve in Europe from the Neander-
thals, where did they come from? In order to answer this question we
need to look at the fossil record of the Middle East.

The Levant: A Crossroads

Although Neanderthals are the result of an evolution which occurred only

in Europe, at some point they left Europe and spread through Central
Asia and the Levant, as demonstrated by the fossils of Teshik-Tash in
Uzbekistan, Shanidar in Iraq, Dederiyeh in Syria, and Kebara, Amud,
and Tabun in Israel.
But in Israel fossils of modern humans have also been found in the
Skhul and Jebel Qafzeh deposits (Figure 14.2).
The Skhul rock shelter lies very close to the Tabun cave. It was excav-
ated by Theodore McCown (1908–69) in 1931–2, and the remains of at
least ten individuals with modern morphology were found there. The
Jebel Qafzeh deposit produced a series of skeletons of at least twenty
individuals. McCown and Sir Arthur Keith studied the Skhul and Tabun
fossils during the 1930s, and concluded that they belonged to a single,
very varied population of Neanderthals. However, the great paleontolo-
gist Francis Clark Howell and Bernard Vandermeersch, director of the
second stage of excavations at Jebel Qafzeh (1961–80), realized that both
the Skhul and the Jebel Qafzeh fossils belonged to modern humans with
archaic traits. We must point out that these modern humans are not
exactly like us; rather they are more like how we might imagine our very
ancient ancestors to look. Some, for example, have marked brow ridges
together with features exclusive to our species, like the chin on the
jawbone, or a high, spherical braincase.

Arsuaga / Chosen Species 1405115327-4-014 Final Proof page 220 4.6.2005 2:44pm

The Fossil Evidence

Figure 14.2 Reconstruction of the head of a woman from the Jebel Qafzeh

Arsuaga / Chosen Species 1405115327-4-014 Final Proof page 221 4.6.2005 2:44pm

The Fossil Evidence

A skull fragment (basically the frontal part of the skull) of uncertain date
found in 1925 in the Zuttiyeh cave, also in Israel, corresponds to an even
older population in this region. But it is not clear whether this is an
ancestor of the Neanderthals or of modern humans, although the latter
seems more probable.
The Neanderthals and the modern humans described here share a
common class of stone technology, the Mousterian, within the great
category of Mode 3 or Middle Paleolithic industries.
We might imagine that the Neanderthals were in the Levant first, and
that modern humans came later and replaced them, as happened in
Europe. There is even the possibility that modern humans evolved
from Neanderthals in the Levant and then migrated to Europe to replace
the Neanderthals there. It is more difficult to imagine that the two human
types were contemporaneous in Israel, and that the Neanderthals lived in
Tabun cave a few minutes away from the modern humans living in the
Skhul rock shelter. This is why it is so important to know the age of both.
The human fossils from Tabun, Skhul, and Jebel Qafzeh have been
dated by the TL, ESR, and uranium series methods, giving an age of
about 100,000 years. However, we need to be more specific in the case of
the Tabun deposit. This cave was excavated between 1929 and 1934 by
the English archeologist Dorothy Garrod, who found a female skeleton
(Tabun 1) and an isolated mandible (Tabun 2). The skeleton is clearly
Neanderthal; the species identification of the mandible (Tabun 2), on the
other hand, is more doubtful, and it might equally belong to a modern
human. Both remains come, in principle, from the level dated at about
100,000 years old. However, Garrod herself admitted the possibility that
the female skeleton was of more recent date and corresponded to the level
above; the people of this upper level might have dug a hole to bury her
and placed her in the lower level.
The Neanderthals of Kebara, Amud, and Dederiyeh are more recent,
between 85,000 and 50,000 years old. These fossils are contemporaneous
with the ‘‘classic’’ European Neanderthals. If the Neanderthal Tabun 1
skull also belonged to this period (which is not certain), and the Tabun 2
mandible was that of a modern human, then all the data would indicate
that the ancestors of modern humans were the first to occupy the region,
as an extension of the settling of Africa; they were then replaced by the
Neanderthals. Archeologist Ofer Bar-Yosef believes that the Neanderthals
colonized the Levant during a period of intense cold in Europe, which saw
a displacement of the populations of central Europe toward the Mediter-

Arsuaga / Chosen Species 1405115327-4-014 Final Proof page 222 4.6.2005 2:44pm

The Fossil Evidence

Finally, between 45,000 and 50,000 years ago the Mode 4 (or Upper
Paleolithic) industries appear in the Middle East, and we may assume that
those who made these tools were the modern humans who, having arrived
from Africa, had replaced the Neanderthals, then spread through the rest
of the world and replaced all other populations of other kinds of humans
they found on their way.

To the Easternmost Edge of Asia

Just as our ancestors reached Finisterre, the westernmost point of Europe,

by replacing the local species, modern humans also spread to the far east of
Eurasia and beyond, reaching Australia more than 40,000 years ago. We
do not know how much earlier, because there is some doubt as to the
dating of the most ancient archeological deposits, but many authors
believe that the first humans set foot in Australia more than 50,000
years ago. One interesting fact is that the Australian deposits contain
Mode 3 tools, but not Mode 4, a fact which supports the hypothesis
that Australia was settled before Europe. Perhaps the ancestors of the
Aboriginal Australians left Africa without passing through Palestine
(where the Neanderthals were at this point), instead crossing the Red
Sea via the Bab-el-Mandeb Strait. They would have had to travel by sea,
but they must have done this in any case to reach Australia, which was
always an island (unlike Java, it had no land connection with the continent
when sea level fell during the glacial stages).
Some of the Australian fossil skulls are very robust, with brow ridges
and even sloping foreheads. Some authors, like Alan Thorne, believe that
these traits indicate that modern Aboriginal Australians carry genes of the
last Indonesian Homo erectus, which the modern humans would have met
on their way and with whom they might have mixed. However, among
the oldest fossils of Australia some, such as those from Lake Mungo
(between 30,000 and 26,000 years old), are of the gracile type, while
others are robust, like specimen WLH 50 (Willandra Lakes), which has
not been dated but could be of similar age. Alan Thorne believes that
the robust and gracile fossils have different origins (with more or less
involvement of Homo erectus), but Peter Brown, another researcher into
the origins of the Australians, believes that both types represent part of the
normal variation of a single population.
In our view, the most likely conclusion is that the Australian Aborigines
have no connection with the populations who lived in East Asia before the

Arsuaga / Chosen Species 1405115327-4-014 Final Proof page 223 4.6.2005 2:44pm

The Fossil Evidence

spread of modern humans. The robust Australian fossils are in fact rem-
iniscent not of Homo erectus but of the first modern humans of Israel and
As in Europe, this spread of modern humans into Asia had fatal conse-
quences for the native populations. In the easternmost area of the world
inhabited by humans, the island of Java, lived the last Homo erectus,
descendants of the populations represented by the Javanese Lower and
Middle Pleistocene fossils. We have a magnificent collection of braincases
of these last Homo erectus, found in the terraces of the Solo River in
Ngandong. As far as is known, in Java as in Europe a local process of
evolution, with a degree of geographic and genetic isolation, took place
here. Morphologically the Ngandong braincases show clear evolutionary
continuity with the older Javanese fossils, although brain size has in-
creased (with cranial capacities ranging from 1,035 to 1,225 cc), modify-
ing the structure of the neurocranium somewhat. The Sambungmacan
braincase is generally considered a good example of a fossil intermediate
between archaic specimens like those of Trinil and Sangiran and the
Ngandong skulls.
It has traditionally been thought, on the basis of the fauna and geology
of the region, that these human fossils from Ngandong belong to the
Upper Pleistocene (less than 127,000 years old), while the Sambungma-
can braincase is usually thought to be older, over 200,000 years old. The
Ngandong Homo erectus would thus be contemporary with the Neander-
thals and modern humans. If this chronology is correct, it would mean
that three different human species coexisted until very recently. Carl
Swisher and his colleagues have used uranium series and ESR dating to
date bovid teeth associated with the human fossils from Ngandong and
Sambungmacan. The result is very surprising, giving a range of ages
between 54,000 and 27,000 years. But once more, we need to wait for
this new dating to be confirmed, given that some eminent geochronolo-
gists, such as Christophe Falguères, disagree and believe the Ngandong
and particularly the Sambungmacan fossils to be considerably older, all
more than 200,000 years old.

The African Origin of Homo sapiens

We have already seen that the first humans who were basically like our-
selves, albeit with some archaic traits, were to be found in the Middle East
(Skhul and Qafzeh) between 90,000 and 120,000 years ago. In South

Arsuaga / Chosen Species 1405115327-4-014 Final Proof page 224 4.6.2005 2:44pm

The Fossil Evidence

Africa there is a deposit of roughly similar age, known as Klasies River

Mouth, which has yielded remains which are fairly fragmented but belong
to modern humans. There are other fossils in Africa with modern charac-
teristics, such as Omo-Kibish 1 and 3 (Ethiopia) and Border Cave (South
Africa); these are believed to be of approximately similar age, although the
dating is less certain. In North Africa fossils with modern characteristics
have been found associated with a local Mode 3 tradition (the Aterian) in
the Dar-es-Soltan II cave (Morocco); their age is uncertain (between
70,000 and 40,000 years old), but they are probably older than the first
settling of Europe by modern humans.
Around one million years ago there is a great gap in the fossil record in
Africa; this gap might well be occupied by a population of the same type as
that represented by the humans of Gran Dolina (Homo antecessor). While
during the subsequent hundreds of thousands of years the European
population of Homo antecessor was to evolve into the Neanderthals, the
African population would evolve into our species (see Figure 12.2).
If this is the case, we need to seek our furthest origins among the
African fossils (of between 600,000 and 250,000 years ago), which
some authors mistakenly include in the species known as Homo heidelber-
gensis (which is actually exclusively European). These fossils are those of
Bodo, Eyasi, Ndutu, Salé, Elandsfontein, and Broken Hill (a group iden-
tified by the name Homo rhodesiensis).
Some fossils from the Asian continent are similar, particularly those
from Dali and Jinniushan. Two fairly complete but deformed skulls
from Yunxian (China) might also be included in this group, as might a
partial skull found in Narmada (India). In theory these populations would
have come from Africa and would have replaced Homo erectus in contin-
ental Asia, perhaps interbreeding with them. In any case, we need to know
more about these fossils and their chronology in order to gain a clear
understanding of what happened in Asia.
Both genetic data (which we shall discuss later) and paleoanthropo-
logical data show that our species has shown a high degree of homogen-
eity in its characteristics ever since it first emerged, indicating that we
derive from a very reduced population which in its turn belonged to a
wider, more varied species. There is a series of human fossils in Africa, less
than 300,000 and more than 10,000 years old, which could already be
qualified as premodern: Omo-Kibish 2 (Ethiopia), Ngaloba 18 (Tanza-
nia), Eliye Springs and ER 3884 (Kenya), Florisbad (South Africa), and
Jebel Irhoud 1 and 2 (Morocco). We ourselves derived from some African
population of this epoch, although we cannot yet say which. The fossils
listed here have, where they are preserved, gracile faces like our own,

Arsuaga / Chosen Species 1405115327-4-014 Final Proof page 225 4.6.2005 2:44pm

The Fossil Evidence

a large brain volume (over 1,350 cc), and braincases which have become
less robust and are rounded in shape (with, for example, less angled
occipital bones than in previous forms). However, as Giorgio Manzi
points out, they have yet to undergo the radical transformation which,
undoubtedly in just one time and place, converted a more or less low
neurocranium with a flat forehead to a rounded, almost spherical neuro-
cranium, widest in the middle and with a high cranial vault and vertical
forehead – the shape we recognize as our own.

Arsuaga / Chosen Species 1405115327-4-015 Final Proof page 226 4.6.2005 2:45pm


The Origins of Modern

Humanity: The Genetic

The value and utility of any experiment are determined by the fitness of
the material to the purpose for which it is used.
Gregor Mendel, Experiments in Hybridization of Plants

A Brilliant Idea

The origin of modern humanity has always been one of the most contro-
versial questions in human paleontology. Many paleontologists believe,
like Gunter Bräuer and Christopher Stringer, that modern humans ori-
ginated in Africa between 300,000 and 100,000 years ago. From this
African birthplace, our species spread through the rest of the Old World
and replaced the other human species (Neanderthals and Homo erectus)
which had appeared as a result of local evolution, in conditions of repro-
ductive isolation, in Europe and Asia. This has been dubbed the ‘‘Out of
Africa’’ hypothesis, in reference to Isak Dinesen’s wonderful book, on
which the film of the same name was based.
Moreover, the idea that Neanderthals and modern humans do not form
a sequence of the ancestor-descendant type, but belong to two independ-
ent evolutionary lines which separated very early on, has also been pro-
mulgated by various anthropologists since it was first put forward in 1912
by Pierre Marcelline Boule (1861–1942).
In the preceding chapters we have followed these approaches to
the origin of modern humanity and its evolutionary relationship with

Arsuaga / Chosen Species 1405115327-4-015 Final Proof page 227 4.6.2005 2:45pm

The Genetic Evidence

the Neanderthals since, in our opinion, the fossil evidence fully confirms
this relationship.
However, it must also be admitted that other paleoanthropologists
find, in the same fossils, evidence of a multiple and very ancient origin
for modern humanity. These researchers believe that each of the different
human populations which occupied the Old World after the first depart-
ure from Africa, over one million years ago, evolved in their separate
geographic regions to give rise to the human populations (the different
‘‘races,’’ to use an older term) which today inhabit the globe.
In its original formulation this hypothesis, championed by Franz Wei-
denreich and Carleton Coon (1904–81) among others, suggested that
each human line had evolved independently and in parallel with the
others. This is not a Darwinist view, since it postulates that different
populations which evolve separately in different environments will ultim-
ately converge into the same species. In order to resolve this problem the
modern version of this theory, whose principal proponents are Milford
Wolpoff and Alan Thorne, suggests that there was a gene flow between all
the Pleistocene populations distributed throughout Africa, Asia, and
Europe. This gene flow, it is suggested, was of sufficient magnitude to
maintain the homogeneity of the human species dispersed through three
continents, but not so intense as to eliminate certain specific traits which
characterize humans in each region. This hypothesis is now known as the
hypothesis of multiregional origin.
The main reason that contradictory theories of the origin of modern
humans continue to exist is the nature of the fossil record. Paleontologists
are trying to unravel a process which took place over hundreds of millions
of years in three continents and involved thousands of individuals. In
order to carry out this task they have no more than a handful of fossils,
often fragmented, isolated, and dispersed in time and space. There is no
doubt that the missing pieces of the jigsaw are more numerous than the
ones we have.
The discovery of new fossils, the precise dating of these, and the ever-
deepening knowledge of the biology of species are the tools paleontolo-
gists use to support their hypotheses. But this procedure is slow and
tortuous, and depends in large measure on the pure chance of paleonto-
logical discoveries. Ideally, we need to have recourse to data from a field
independent of paleontology in order to put the hypotheses based on the
fossils to the test. But where can we turn to find such data?
The knowledge gained over the 20th century about the mechanisms of
genetic inheritance allows us to consider a new way of approaching the
problem of the evolutionary history of species. The idea is as simple as it is

Arsuaga / Chosen Species 1405115327-4-015 Final Proof page 228 4.6.2005 2:45pm

The Genetic Evidence

brilliant: it does not matter that there are few fossils of past species, since
the genetic material in living species contains the key to their evolutionary
history. We just need to know where to look.
We have already seen an example of this approach in the chapter on the
origin of hominids and their relation to gorillas and chimpanzees. But the
problem of the origin of modern humanity is different because only one of
the species involved in the process survives, and we therefore only have
genetic material from one of them (with one exception, which we shall
look at later in this chapter). Genetic studies dedicated to clarifying the
origin of modern humanity attempt to determine the genetic structure of
humans today; on this basis inferences can be drawn about the how, when,
and where of our origin.

The Molecules of Inheritance

The molecule responsible for biological inheritance is deoxyribonucleic

acid (DNA), which carries, coded in its chemical structure, the informa-
tion necessary to ensure the continuation of the species. DNA in the cells
is organized into discrete units known as chromosomes, which in animals
are housed in the nucleus of the cell. Each species has a specific number of
chromosomes: humans have 23 different pairs of identical chromosomes,
46 in total.
When gametes (ova and sperm) are produced, a very special kind of
cellular division known as meiosis occurs, as a result of which each gamete
receives only one copy of each chromosome. In other words, unlike the
other cells in the body, our gametes have only 23 chromosomes, one of
each type. The significance of this process is clear: if the gametes carried
the same number of chromosomes as the rest of the organism’s cells, the
egg cell which results from the fusion of two gametes would have double
the number of chromosomes as the cells of its progenitors, and the
number of chromosomes in a species would not remain constant over
time but would double with each generation.
During meiosis a very important phenomenon known as recombination
occurs; this consists in the exchange of fragments of DNA between the
homologous chromosomes of each pair. As a result two recombinant
chromosomes are obtained, whose genetic information is a mixture of
that of the paternal and maternal chromosomes. The phenomenon of
recombination is extremely valuable in evolutionary terms, since mixing
the information received from each progenitor results in the appearance of

Arsuaga / Chosen Species 1405115327-4-015 Final Proof page 229 4.6.2005 2:45pm

The Genetic Evidence

new genetic combinations. Together with mutation, recombination is the

main factor responsible for variation in organisms, the base on which
natural selection operates.
The 23rd chromosome pair is different from the others because in
males it is formed by two different chromosomes. It is this pair of
chromosomes that holds the genetic information determining the sex
of individuals. Chromosomes in pair 23 can be of two types: one an
X-shaped chromosome (giving it its name, X-chromosome), and the
other smaller, Y-shaped chromosome (known as the Y-chromosome).
Women always have two X-chromosomes in the 23rd pair, while men
have one X- and one Y-chromosome; thus women are XX and men XY.

Black Eve

Our cells obtain their energy from a series of very complex chemical
reactions, most of which, particularly those which involve oxygen, take
place within a series of small organelles known as mitochondria. In add-
ition to performing this important role in cell life, the mitochondria are
exceptional in another way: they are the only organelles in the animal cell
which have their own genetic material.
The DNA of a mitochondrion is contained in a circular chromosome
smaller than the chromosomes of the cell nucleus, and very similar to that
of bacteria. Mitochondrial DNA is ideal for evolutionary studies for two
reasons: first because all variation is due exclusively to mutation, since it is
not subject to recombination, and secondly because the organelles in the
egg cell come only from the maternal ovum and are transmitted through
the mother’s line (in the process of fertilization the sperm only contrib-
utes its nuclear chromosomes, so the egg cell consists of the ovum plus the
nuclear chromosomes of the sperm).
Thus, we can trace the ancestry of a mitochondrial chromosome, from
woman to woman, over the generations. The mitochondrial DNA (for
which we shall use the standard abbreviation, mtDNA) in any of our cells
can be identified with a single ancestor in each generation: our mother,
our maternal grandmother, only one of our four great-grandmothers (the
mother of our grandmother), and so on.
One might naı̈vely suppose that geneticists analyze the entire mtDNA
of each individual to identify its peculiarities in order to compare them
with those of others, but such a task would require a great deal of time,
resources, and effort. Studies on variation in mtDNA are actually confined

Arsuaga / Chosen Species 1405115327-4-015 Final Proof page 230 4.6.2005 2:45pm

The Genetic Evidence

to specific regions. The regions chosen must show variation, manifested in

the existence of a series of different types of mtDNA (polymorphism). The
various modern human populations can be characterized in terms of the
frequency in which the corresponding types occur in each of them.
Although it was not the first work published on variation in mtDNA in
humans, the article by Rebecca Cann, Mark Stoneking, and Allan Wilson
which appeared in Nature on January 1, 1987 caused a major upset in the
field of studies of the origin of modern humanity. The article presented
the results of an extensive study of the mtDNA of 147 individuals from
5 large groups of different types of humans (Caucasians, Asians, Africans,
Australian Aborigines, and natives of Papua New Guinea). The breadth of
the sample, together with the extent of the portion of mtDNA analyzed
(representing about 9 percent of the total mitochondrial chromosome)
contributed to the impact of the article.
The results of this study can be summed up in two main points (with
which other, more limited studies, made previously by other authors,
First, on the basis of similarity of mtDNA, the existence of two major
groups was noted. In one of these only mtDNA of African origin is found,
while in the other mtDNA from the other four groups appears together
with some mtDNA of African origin.
The second fundamental result of the study relates to the variation
within each group. The mtDNA of the African group showed greater
diversity than that of the group which included the rest of the mtDNA.
This was interpreted as evidence that the African group was the oldest of
all. The use of the variation in mtDNA found within a given group as a
measure of its antiquity is based on the assumption that the older the
group, the more time it will have had to accumulate mutations, giving rise
to more different types of mtDNA.
Finally, the authors of the article calculated the date at which the
mtDNA lines separated as about 200,000 years ago, the time when the
woman to whom the two lines could be traced lived in Africa.
Cann, Stoneking, and Wilson’s conclusions hit the headlines in the
mass media and were immediately dubbed the Black Eve hypothesis (an
allusion to the African origin of our species); they were also immediately
The main objections to Cann and her colleagues’ hypothesis related to
the interpretation of their results and their method of estimating time.
There are two arguments criticizing the interpretation of the results.
The first stresses the fact that, by the very nature of its matrilineal trans-
mission, we might expect lines of mtDNA to be lost over time simply as a

Arsuaga / Chosen Species 1405115327-4-015 Final Proof page 231 4.6.2005 2:45pm

The Genetic Evidence

result of chance (for example, the mtDNA of women who only give birth
to males will cease to be represented in the population); this alone would
be enough to explain the limited variation in non-African populations.
Another criticism is based on the argument that mtDNA studies may
offer a biased vision of the evolutionary history of humans, since they
consider only the history of women, which might not be the same as that
of the overall population.
These objections have met responses and counter-arguments, drawing
in a large number of scientists. Since all the arguments put forward are
reasonable, the only way to resolve the problem is to look for new
evidence by studying nuclear DNA.
We will return to the issue of calculating the time elapsed since the
origin of modern humans later in this chapter.

An Adam for Eve

The best way of comparing results and interpretations based on the study
of mtDNA is by studying the variation of a part of the nuclear DNA which
is transmitted paternally and which, like the mitochondrial chromosome,
is not subject to recombination. The only nuclear chromosome fulfilling
these criteria is the Y-chromosome.
The variation in some of the polymorphisms identified in the
Y-chromosome can be categorized as one of a few types (or haplotypes)
among which we can determine which is the most primitive (by compar-
ing them with the condition found in modern apes). This is a valuable
innovation with respect to earlier studies of the mitochondrial chromo-
some, in which the primitive type was identified on the basis of
the distribution of different types of mtDNA among the populations
The results of various analyses of different polymorphisms of the
Y-chromosome all point in the same direction: modern humans had a
male ancestor who lived in Africa between 100,000 and 200,000 years
ago. The two most recent studies, carried out by teams led by Michael
Hammer and Peter Underhill, go even further, indicating that the
Khoisan people (the Bushmen) are the human population with the
highest frequency of primitive haplotypes.
But in addition to these results relating to the origin of our own species,
the Y-chromosome studies have given us very valuable information on
other aspects of our evolutionary history. First, that there was no one

Arsuaga / Chosen Species 1405115327-4-015 Final Proof page 232 4.6.2005 2:45pm

The Genetic Evidence

single ‘‘wave’’ of migration from Africa, but at least two, occurring at

different times. The first of these was over 50,000 years ago and resulted
in the colonization of Asia and Australia; another, later ‘‘wave’’ arrived in
Europe. These results match up with the archeological data which, as we
have seen, indicate that Australia was settled by humans before Europe.
But perhaps the most striking result of comparison of the mtDNA and
Y-chromosome studies is that, while the different variants of mtDNA are
widespread throughout the world, the different types of Y-chromosome
show more limited geographical distribution, and many of them are
restricted to local groups. Luigi Cavalli-Sforza suggests a bold interpret-
ation of these data: that it was women who spread their genes throughout
the world, while the men stayed preferably in the birth group; thus,
societies were patrilocal. It is worth remembering here Rob Foley’s hy-
pothesis (discussed in the chapter on social biology) that the first hom-
inids, like chimpanzees, formed patrilocal societies of related males.

The Other Chromosomes

Although the analyses of the mitochondrial and Y-chromosomes lead to

similar conclusions, it can be argued that these results are based on studies
limited to a small part of a person’s DNA and that, in addition, they are
restricted to very specific chromosomes, given their particular link to one
or other sex. We can legitimately question whether the ‘‘evolutionary
history’’ of the other chromosomes would also support African origin or
whether, on the other hand, it would suggest a broader view.
James Wainscoat was a pioneer of the study of the origin of modern
humanity on the basis of nuclear DNA. Working on the distribution of
five polymorphisms in the region of the hemoglobin gene for eight
human groups, in 1986 Wainscoat and his team announced that all
modern human populations derived from an ancestral African population
around 100,000 years ago, which would have numbered about 600
individuals. The conclusions published by Luigi Cavalli-Sforza and his
colleagues in 1988, on the basis of analysis of the distribution of 120
genetic markers (proteins coded by nuclear DNA, such as blood groups)
in 42 human populations, were along the same lines. The African origin
of modern humanity about 100,000 years ago was corroborated once
again in 1991 by a broad study of polymorphisms in nuclear DNA carried
out by two teams, led by Luigi Cavalli-Sforza and by Judith and
Kenneth Kidd.

Arsuaga / Chosen Species 1405115327-4-015 Final Proof page 233 4.6.2005 2:45pm

The Genetic Evidence

Pleistocene Park

Michael Crichton’s novel Jurassic Park, the starting point for Steven
Spielberg’s film series, is based on the possibility of recovering intact
DNA from fossils. When Crichton wrote his book, a series of scientific
articles reporting on the discovery of dinosaur DNA through insects
fossilized in amber around 100 million years ago had just been published.
However, these studies have now been completely discredited and it has
been recognized that the DNA found was due to modern-day contamin-
ation. Moreover, it has been demonstrated that it is impossible to find
fossil DNA of such an age, for the simple reason that the DNA molecule
cannot survive unaltered for this length of time; not even amber can
prevent the oxidation of DNA and its consequent deterioration.
The oldest DNA which has been recovered from fossils is much more
recent than that of the dinosaurs in Jurassic Park; it is mtDNA from
mammoths preserved in the ice of Siberia between 50,000 and 100,000
years ago. The low temperatures seem to have favored the preservation of
the mammoths’ mtDNA, and there is no expectation that anything similar
will be found outside of very specific environments such as the Siberian
One of the main problems for the ‘‘DNA paleontologist’’ (apart from
that of whether fossil DNA actually exists) is avoiding contamination with
modern DNA through the manipulation inherent in the excavation,
restoration, study, and other work carried out on fossils. The efforts
made to isolate mtDNA from the famous Ice Man of the Tirol (5,000
years old) offer a good example. The first attempts were thwarted by the
presence of contaminating modern DNA, and the techniques had to be
refined in order to find mtDNA whose authenticity was not in doubt.
Although the results obtained from the study of the mtDNA from the
Ice Man were modest (his mtDNA was shown to belong to a type
characteristic of central European populations), the study did offer some
very valuable lessons on the techniques and controls necessary in order to
get around the problem of contamination with modern human DNA.
Using these new procedures, an international team (comprising Mat-
thias Krings, Anne Stone, Ralph Schmitz, Heike Krainitzki, Mark Stone-
king, and Svante Pääbo) embarked on the search for mtDNA in
Neanderthal fossils. They took samples from the type specimen of this
species, the Neanderthaler skeleton. The study was very carefully planned
and surrounded with all possible precautions, both to avoid contamin-
ation as far as possible and to prevent pointless destruction of a fragment

Arsuaga / Chosen Species 1405115327-4-015 Final Proof page 234 4.6.2005 2:45pm

The Genetic Evidence

of this valuable fossil. The 3.5-g sample was taken from a place theoret-
ically inaccessible to contamination, the inner part of one of the bones –
specifically, the right humerus.
The next step in the study was to analyze the structure of the fragments
of what was presumed to be Neanderthal mtDNA. The DNA molecule is
made up of thousands of smaller units known as nucleotides, of which
there are only four different types in DNA. The structure of a fragment of
DNA is simply the sequence of nucleotides which make it up, so this part
of the work consisted of determining the nucleotide sequence in the
different fragments of DNA found in the sample taken from the fossil.1
Many of these fragments showed sequences which were identical over
some part of their length: in other words, they overlapped. This made it
possible to reconstruct the sequence of the original segment from which
they came. Thus, after three months of intensive work, the sequence of a
segment of 379 nucleotides was reconstructed, on the basis of 123
different fragments. This sequence corresponds to region I of the control
segment of mtDNA.
The researchers compared the mtDNA sequence from the fossil with 16
types of mtDNA from chimpanzees and 1,986 types from different mod-
ern human populations, and obtained some extremely valuable results.
First, they determined that on average humans and chimpanzees differ in
55 bases of the base sequence, while the mtDNA analyzed showed an
average difference of 27 bases from modern humans. Moreover, they
established that within the modern human sample the average difference
was only 8 bases. Thus, the fossil mtDNA is sufficiently similar to our own
to recognize that it comes from a human being, but different enough to
reject the suggestion that it belongs to a modern human, thus ruling out
the possibility of contamination and confirming it as an authentic frag-
ment of Neanderthal mtDNA.
The fact that the average difference between Neanderthal mtDNA and
that of modern humans is more than three times greater than the average
difference among modern humans (27 bases compared to 8) led the

A series of fragments of mtDNA, representing two types of sequence, was obtained
from the Neanderthal fossil. One of these sequence types was attributed to modern
mtDNA, present through contamination, while it was thought that the other might
correspond to the authentic mtDNA of the fossil. It is striking that despite all the
precautions taken, which theoretically rendered the presence of contaminating DNA
impossible, it did appear in the analyses. This fact indicates how difficult it is to avoid
contamination in this type of work, and throws into question the results of other, less
rigorous studies.

Arsuaga / Chosen Species 1405115327-4-015 Final Proof page 235 4.6.2005 2:45pm

The Genetic Evidence

researchers to conclude that the two lineages separated a very long time
ago.2 In order to establish when this separation occurred, the researchers
used a mutation rate determined on the basis of the average number of
changes observed between modern humans and chimpanzees (55) and
the time which is assumed to have passed since the two lines separated.
Their calculations gave an age of between 550,000 and 690,000 years for
the Neanderthal/modern human separation, and between 120,000 and
150,000 years for the origin of present-day human diversity.
The conclusions about the age of modern humans concur with those
derived from studies of mtDNA, the Y-chromosome, and other nuclear
chromosomes. One secondary result of the analysis is that the mtDNA of
modern humans of African origin again appears ancestral to that of the
other modern human populations.
In addition to reinforcing the theory of a single, African origin for
modern humanity, this study has another noteworthy feature of particular
relevance. The separation between Neanderthals and modern humans is
traced much further back than the majority of authors working on fossil
studies had suggested up to this point. However, this separation date is
compatible with the results of our investigations of the human fossils of
Atapuerca, both from La Sima de los Huesos and from Gran Dolina level
6. As we have already noted, the Gran Dolina fossils, which are close to
800,000 years old, represent the ancestor species of both the Neanderthal
and the modern human lines, Homo antecessor.
Perhaps the discovery of Neanderthal fossil mtDNA will give rise to a
new novel, a tale of cloning Neanderthal individuals. However, one of the
conclusions which can be drawn from our better knowledge of fossil DNA
is that we will never find sufficient nuclear DNA to allow us to dream of
the Pleistocene version of Crichton’s book.

Fossils and Molecules

It has frequently been suggested in the popular media that genetic studies
just by themselves have revealed the ‘‘how,’’ ‘‘when,’’ and ‘‘where’’ of our

Nevertheless, some comparisons of mtDNA within modern humans gave a greater
average difference (up to 24 bases) than that obtained from comparison of Neanderthal
mtDNA with that of certain modern humans (20 bases). Thus the results of the study
are not as convincing as has been suggested, but should rather be considered statistic-
ally reliable.

Arsuaga / Chosen Species 1405115327-4-015 Final Proof page 236 4.6.2005 2:45pm

The Genetic Evidence

origin. However, as we pointed out at the beginning of this chapter, the

main hypotheses as to ‘‘when’’ and ‘‘where’’ have been based on the study
of fossils, and were put forward before the genetic studies began.
In this context, the genetic studies have reinforced the hypothesis of the
African origin of modern humans (the ‘‘Out of Africa’’ hypothesis).
Moreover, genetic analyses have also corroborated the opinion of many
paleontologists that Neanderthals were not either direct or indirect evo-
lutionary ancestors of modern humans, but that the two human species
share, far back in time, a common past.
One important area which genetic studies have been essential in illu-
minating is the ‘‘how’’ of our origin. The narrow genetic range noted
in non-African human populations indicates the existence of an evolu-
tionary phenomenon known as a ‘‘bottleneck’’ at the point when our
species left Africa. ‘‘Bottlenecks’’ are created when a biological population
originates from a relatively small number of individuals who carry only a
fraction of the genetic diversity of the mother population. As a result,
the population derived from this small group receives only a part of this
We can see a good example of this phenomenon in the Afrikaner
population of South Africa, which largely originates from a handful of
Dutch pioneers who arrived there in the 17th century. There is evidence
that one of those colonists, who arrived in 1688, suffered from a rare
genetic disease known as porphyria. And we find that today, the frequency
of this disease in the South African Afrikaner population is several hundred
times higher than in any other human population.
Similarly, the high degree of homogeneity of the non-African popula-
tions tells us that the modern human groups which colonized Asia and
Europe were formed by a small fraction of the original African population.
The number of pioneers can also be fairly reliably established. A team of
geneticists including Spanish-born researcher Francisco Ayala have ana-
lyzed the present-day variation, in modern humans, in the genes respon-
sible for the main histocompatibility complex, or HLA system, which is
involved in defense against microbial invasion through its capacity for
recognizing proteins alien to the organism.
These genes are located on chromosome 6 and are extremely varied,
allowing the HLA to recognize a large number of molecules as alien. The
range of variation observed in modern humans indicates that the ‘‘bottle-
neck’’ through which our ancestors passed was not too tight, since other-
wise there would be much less variation in the HLA system. According to
these researchers, the number of settlers who left Africa must have been
greater than 500, and was probably around 10,000.

Arsuaga / Chosen Species 1405115327-4-015 Final Proof page 237 4.6.2005 2:45pm

The Genetic Evidence

While genetic studies have been useful in comparative analysis of pale-

ontological hypotheses on the ‘‘where’’ of the origin of modern humans,
and have been particularly valuable as a source of information on the
‘‘how,’’ they have not been so eloquent on the question of when we
The measurements of time calculated in the genetic studies, known as
‘‘molecular clocks,’’ are always problematic because they are based on a
series of highly debatable assumptions, which we have already discussed in
the chapter on the origin of hominids (the assumption that the rate of
mutation is constant, or that the regions of DNA studied are neutral with
regard to the action of natural selection).
Perhaps the most serious criticisms of ‘‘molecular clocks’’ are those
which cast doubt on the rates of mutation used to calculate the moment
when modern humans emerged. In their original article Cann, Stoneking,
and Wilson recognized that time elapsed cannot be reliably established on
the basis of variation in mtDNA alone. Like all other authors, they adjust
their ‘‘molecular clocks’’ by using the fossil record, and therefore their
results cannot be invoked to compare hypotheses based on fossils. In this
respect paleontology and genetics are not independent sources.
At the beginning of the 21st century, our view of the origin of modern
humanity is much more complete than might have been expected even 20
years ago. We are much more certain of the place and time when our
species emerged, and we have made a great deal of progress in clarifying
how the process took place. Contrary to some expectations, molecules
have not replaced fossils in the study of human evolution. The different
approaches of paleontology and genetics have enabled us to consider the
problems from different perspectives, enriching our knowledge of the
evolutionary history of the species Homo sapiens.

Standards of Beauty

Up to now we have been considering our species’ past, but what of its
future? Many believe that, now we have machines enabling us to depend
less and less on our physical strength to survive, and more and more on
our intelligence, the corresponding organs will be affected in the future.
Thus the human being of tomorrow is frequently represented as having an
atrophied body and a very big head, or rather, with an overdeveloped
brain, since the face and teeth are also depicted as reduced – in short, a
very unattractive individual, in terms of the ancient Greek canons of

Arsuaga / Chosen Species 1405115327-4-015 Final Proof page 238 4.6.2005 2:45pm

The Genetic Evidence

beauty. These theories usually give no precise date as to when we will

become such unathletic, although highly intelligent creatures, but it
seems that at the rate we are going the standards of beauty will have to
change very quickly (goodbye to 36–24–36!).
But is it possible that natural selection is still acting on us, and can
determine the future course of human evolution? There is absolutely no
doubt that our species is subject to natural selection, like all others.
Individuals with serious genetic defects do not live to adulthood and do
not reproduce, and many die in utero and are not even born. But this
normalizing selection, which eliminates extreme individuals, does not
modify the species. For the species to evolve in a given direction takes a
very long time, and it also requires individuals with specific characteristics
to reproduce more than others, something which appears not to be
happening, or not on any great scale. Moreover, technology allows us to
adapt rapidly to life in all kinds of environments, including the moon,
without changing our morphology. Adaptation by natural selection is
much slower (and more limited). To take just one example, thanks first
to writing and now to information technology, our brain does not need to
grow any more in order to accumulate and process more information.
On the other hand, our species is already very numerous, and there is
therefore a great genetic inertia or resistance to changes, which are diluted
like drops of water in the ocean. An interesting phenomenon is beginning
to arise, which will go against the trajectory of humanity over the last few
thousand years: human populations, which became isolated from one
another and evolved into different races, are beginning to mix and ex-
change genes, and it is thus certain that there will be new genetic com-
binations. But this does not mean that the species will change substantially
in the near future.
One final, disturbing factor should be mentioned. Since we discovered
artificial selection 10,000 years ago with agriculture and animal hus-
bandry, we have always been capable of modifying ourselves just as we
have modified breeds of animals. It does not appear that this has happened
on any major scale. Now, however, with our knowledge of genetics, we are
beginning to have the real possibility of modifying our own genes much
more rapidly and radically than with artificial selection (and even more
rapidly and radically than natural selection). Genetic manipulation, which
can free us from defects and disease, could also be used for other purposes.
But in itself it is simply a tool, like all those that science makes available to
us, and it is our responsibility to control it.

Arsuaga / Chosen Species 1405115327-4-016 Final Proof page 239 8.6.2005 5:24pm


The Origins of Human


Thus it was necessary for the said hole in the Adam’s apple to close when
meat is swallowed that it does not enter the throat, which would be a
vexing and harmful thing, of which we have experience every day when
we hurry in swallowing and something enters our throat, because there
follows from this a most troublesome and grievous coughing.
Bernardino Montaña de Monserrate, Libro de la anathomia del
hombre [Book of the Anatomy of Man]

King Solomon’s Ring

Human beings are the only organisms which speak. That is to say, we
transmit to our fellows, and receive from them, all kinds of new informa-
tion, deliberately encoding our messages in combinations (words) of
pre-established sounds (syllables). Other animals are only capable of
exchanging very specific information on certain aspects of their life,
using a limited system of sounds and gestures which are not intentionally
Konrad Lorenz, in a delightful book entitled King Solomon’s Ring,
makes reference to the legend that King Solomon possessed a ring
which enabled him to talk with beasts. Lorenz boasted that he too was
able to understand the simple vocabulary of animals, without needing a

Arsuaga / Chosen Species 1405115327-4-016 Final Proof page 240 8.6.2005 5:24pm

The Origins of Human Language

ring; but he added that animals do not have a true language; rather, each
individual innately possesses a code of signals formed by expressive sounds
and movements of which another individual of the same species has a
similarly innate understanding. However – and this is where they differ
fundamentally from humans – animals emit these signals automatically
when they find themselves in a particular state of being, even if there is no
one to witness them. Lorenz expressed this idea by saying that with their
sounds animals are emitting not ‘‘words’’ but ‘‘interjections.’’ The
Merriam-Webster Dictionary defines an interjection as ‘‘a cry or inarticu-
late utterance expressing an emotion.’’ And this is exactly what animals are
emitting, according to Lorenz.
These ideas appear to be true for the majority of animals, but they are
perhaps not quite so accurate in the case of primates. Dorothy Cheney and
Robert Seyfarth have studied vervet monkeys (Cercopithecus aethiops) in
the wild in Africa, observing that in addition to emitting signals, in the
form of sounds and gestures, which communicate their motivation or
state of mind, these animals also inform one another about certain aspects
of the environment. For example, they have different sounds to warn of
the presence of different predators – snake, eagle, or leopard. The reac-
tions that these calls stimulate in the hearers are different in each case: if
the warning is of a leopard they climb trees; if the predator is an eagle they
hide in the bush or look up; and if a snake is signaled they stand up and
scan the grass. Thus, each call has a distinct meaning which triggers a
different response; these are not simply cries of fear in reaction to the
presence of a predator.
Moreover, Cheney and Seyfarth’s research on other calls related to the
social life of the vervet monkey has revealed that these monkeys associate
sounds which have similar meanings, even if they are acoustically very
different (just as we would with the words ‘‘car’’ and ‘‘vehicle’’).
In the 1960s and 1970s the idea of direct communication with the
animals most like us, chimpanzees and gorillas, was taken very seriously in
some research programs. Since chimpanzees and gorillas are physically
unable to pronounce words, the task of communicating with us was
facilitated by teaching them sign language, which they were able to
reproduce. The chimpanzees and gorillas proved to be outstanding pupils,
and revealed their capacity for associating ideas which we express in words
with gestures, or with tokens of different shapes, with drawings and colors
The bonobo Kanzi (whose efforts at carving stone tools we described in
a previous chapter) understands more than 150 words of spoken English.
And Kanzi was not the only chimpanzee to show a degree of linguistic

Arsuaga / Chosen Species 1405115327-4-016 Final Proof page 241 8.6.2005 5:24pm

The Origins of Human Language

capacity. Washoe was the first chimpanzee to learn a series of sign-lan-

guage gestures (132 signs during a little over 4 years of training); Sarah
showed indications of an ability to detect the order of objects used as
words; and Lana appropriately completed sentences made up of characters
representing words, paying attention to the meaning and order of the
One classic question relating to animals’ systems of communication is
whether they can deceive one another for their own benefit. Knowing
how to lie would make them more ‘‘human,’’ as it would indicate that
they are not simply automatons, but are able to control what they express.
In fact, chimpanzees in the wild have been observed to deceive their
companions on many occasions, in a wide variety of contexts, with their
gestures, posture, and facial expression. Our sins are also theirs.
The results of all these investigations are extremely valuable, because
they have revealed an incipient linguistic skill in primates which was
previously denied. However, they are disappointing in that none of
these primates has communicated any relevant information on themselves.
Vervet monkeys appear to have a limited repertoire of ‘‘words’’ which
they use in very specific situations, and chimpanzees have demonstrated
that they are very skillful in manipulating symbols, and can even be
consummate liars, but that is all. Perhaps the myth of King Solomon’s
ring is, after all, just that – a myth.
Given that human language is so different from that of our living
relatives, the question of its origin and development can only be tackled
from the viewpoint of paleontology.

Language and Brain

Paleoneurology is a science which attempts to determine the mental cap-

acities of a fossil species from the impressions left by the brain on the inner
surface of the skull. There are two areas of the cerebral cortex, both in the
left hemisphere of the brain, that are closely related with speech in humans
(see Figure 8.1). Broca’s area, located in the third frontal gyrus (at the
level of the temple), is responsible for syntactic construction and planning:
in other words, it translates messages into an ordered sequence of move-
ments of the muscles involved in the production of speech. A lesion in this
area disrupts the capacity to speak and write, but not the individual’s
understanding of spoken language, and the affected person will still be
able to read. Wernicke’s area, on the other hand, located between the

Arsuaga / Chosen Species 1405115327-4-016 Final Proof page 242 8.6.2005 5:24pm

The Origins of Human Language

upper temporal gyrus and the parietal lobe (a little behind and above the
ear), is responsible for encoding and decoding messages. A lesion in
Wernicke’s area renders the individual incapable of correctly understand-
ing and producing language, whether spoken or written.
According to Philip Tobias, the lower region of the parietal lobe, the
part related to Wernicke’s area, is more developed in the Homo habilis
fossils from Olduvai than in australopithecines, Paranthropus, and apes.
Moreover, Broca’s area is markedly enlarged both in Homo habilis/Homo
rudolfensis and in Homo ergaster. This area is much more developed in the
first humans than in australopithecines and Paranthropus, in which it is
only sketchily present.
Thus, the regions of the cerebral cortex most directly related to the
production of human language were already well developed in the first
representatives of our genus. Does this mean that these humans already
possessed the capacity of speech? Although this is the conclusion of the
majority of specialists who have studied brain casts from primitive hom-
inids, there are those who oppose this view.
In his book The Wisdom of the Bones, Alan Walker, who led the study of
the Turkana Boy, rejects the possibility that this individual, and by exten-
sion all those of his species, was able to speak. Walker bases his conclusion
on the fact that within the thoracic vertebrae of Turkana Boy the vertebral
canal is very narrow. This configuration is common among apes, but not
in modern humans, who have a much wider vertebral canal. In view of the
reduced diameter of the vertebral canal, Walker argues that the spinal cord
of the fossil specimen contained fewer neurons than that of modern
humans, so Turkana Boy’s thoracic region would have been less innerv-
ated than our own. The only plausible explanation of this fact, according
to Walker, is that the thoracic musculature involved in breathing was not
capable of performing the precise inhalations and exhalations that human
speech requires. So how do we explain the high level of development of
Broca’s area reflected in the inner surface of the braincase of Turkana Boy?
On the basis of results obtained from modern techniques used to explore
brain activity (specifically positron emission tomography, or PET), which
show that the area of the cerebral cortex surrounding Broca’s area is also
related to manipulation of the right hand, Walker suggests that the
development of this area in the first humans was an adaptation related
not to speech, but to dressing stone.
To sum up, although different studies of the cerebral cortex of the
first hominids concur that the areas connected with language (particularly
Broca’s area) are more fully developed in the first humans than in austra-
lopithecines, Paranthropus, and apes, they do not agree on the

Arsuaga / Chosen Species 1405115327-4-016 Final Proof page 243 8.6.2005 5:24pm

The Origins of Human Language

physiological significance of this development. The solution to this prob-

lem could come from research into the anatomy of the phonetic apparatus
of fossil hominids.

The Choking Primate

The second route of investigation into the origins of language is to study

the set of organs responsible for the emission of the sounds that make up
human speech. Although this might at first appear to be of secondary
importance, given that the mental capacities required to possess a complex
language do not depend on the physiological capacity for producing it, it
is nevertheless true that one cannot compose music for instruments that
do not exist.
The sounds on which human language is based are produced and
modulated in a series of cavities which make up the upper stretch of the
respiratory tract and are known collectively as the vocal tract: the larynx,
pharynx, and the nasal and oral cavities (Figure 16.1).
In all mammals apart from adult humans, the larynx is situated high in
the neck, almost at the exit of the oral cavity. This high position allows the
larynx to connect with the nasal cavity during the ingestion of liquids,
which thus pass from the oral cavity to the alimentary canal without any
interruption of breathing. In other words, any mammal is able to breathe
through its nose while drinking. However, in adult humans the larynx is
positioned unusually low in the neck, so that although we are mammals,
we are unable to breathe while drinking.
The importance to a mammal of being able to breathe through the nose
while drinking is clear if we think of young creatures who are being
breastfed. For this system of feeding to be efficient, an infant must be
able to breathe as it suckles.
It will not have escaped the perceptive reader that our babies are also
able to breathe through the nose while suckling or drinking from a bottle.
This shrewd reader will thus have correctly surmised that in nursing
humans the larynx is in the same position as in other mammals (Figure
16.2). In our species the larynx descends at the age of about two years.
From this point, not only do we lose the capacity to breathe while
drinking, but the unusual position of the human larynx makes it possible
for food to obstruct the respiratory tract, since the epiglottis is not able to
cover it completely (Figure 16.1). Choking is no joke – we can die from it.

Arsuaga / Chosen Species 1405115327-4-016 Final Proof page 244 8.6.2005 5:24pm

The Origins of Human Language

But if our upper respiratory tract has lost efficacy in this respect (and also
for breathing and sense of smell), what is the compensation? The answer is
that our species has a pharynx larger than that of any other mammal, giving
us the capacity to modulate a wide range of different sounds.

The Production of Speech

Contrary to popular belief, the majority of the basic sounds which make
up human speech do not originate directly in the vocal cords. In the
production of vowel sounds (and also of voiced consonants, but for the
sake of simplicity we will refer only to vowels in what follows), the vocal
cords open and close rapidly with periodic breaths of air to produce a
‘‘base’’ sound, or laryngeal tone. This is always the same, regardless of the
vowel we are pronouncing.
The laryngeal tone is made up of one main frequency and a series of
‘‘accompanying’’ frequencies or harmonics. If the set of cavities above
the larynx (the pharynx, nasal cavity, and oral cavity) were not involved


Nasal cavity

Oral cavity

Location of
vocal cords and

Alimentary canal
Hyoid bone
Thyroid cartilage
Respiratory tract

Figure 16.1 (cont’d)

Arsuaga / Chosen Species 1405115327-4-016 Final Proof page 245 8.6.2005 5:24pm

The Origins of Human Language



Location of vocal
Hyoid bone cords and glottis

Thyroid cartilage

Figure 16.1 Section of the head of a common chimpanzee and a human,

showing the upper airways. On its inward passage towards the lungs, air usually
enters via the nasal cavity. From here it passes to the pharynx, a channel common to
both the alimentary canal and the respiratory tract, serving as a passage for both air
and food and fluids. The pharynx continues downward to the point where the
alimentary canal and respiratory tract separate. The entrance to the respiratory tract
is formed by a cartilaginous vessel known as the larynx. The upper part of the larynx
is formed by the thyroid cartilage, which forms an easily distinguished protuberance
in the neck (particularly in men), the ‘‘Adam’s apple.’’ In the lower part of the
larynx there are two bands of muscle covered with an elastic sheath, the vocal cords.
The space between the vocal cords is called the glottis. The primary function of the
vocal cords is to block the glottis when they close together, thus preventing foreign
bodies entering the respiratory tract. The epiglottis, a spoon-shaped cartilage
located above the larynx, also contributes to this function. When we swallow or
drink, the larynx rises so as to position the glottis beneath the epiglottis, partially
blocking the respiratory tract against the passage of food and fluids. As a result,
breathing is interrupted for the duration of the time the larynx remains closed.

Arsuaga / Chosen Species 1405115327-4-016 Final Proof page 246 8.6.2005 5:24pm

The Origins of Human Language

in the production of human speech sounds, the sound we heard would

consist mainly of the note corresponding to the principal frequency
of the laryngeal tone; most of its harmonics are too weak for us to
hear them.
However, this is not what actually happens, owing to the phenomenon
known as resonance, whereby a body (resonator) is able to produce
vibrations as a result of the vibration of another adjacent body. This is
why window glass vibrates (resonates) in response to traffic noise. How-
ever, not all bodies which act as resonators reproduce the same vibrations:
each type of resonator responds to certain particular vibrations. Thus if we
pass a sound made up of various frequencies (as the laryngeal tone is)
through a resonator, the resonator will reproduce, and even amplify, some
frequencies but not others. The result will be an acoustic filter, where
some frequencies have been amplified and others silenced; the resulting
sound will be different from the original. Depending on the resonator we
use, we will get different sounds from any given base sound.
The human vocal tract can adopt different configurations, each of
which acts as a different resonator which filters the laryngeal tone pro-
duced by the vocal cords in a specific way, giving rise to the different vowel
sounds. In order to generate this filter, narrowing and widening has to
take place at specific distances from the source of the laryngeal tone, both
in the pharynx and in the oral cavity.
For example, in order to produce the sound [a], the tongue flattens in
the oral cavity and shifts backward to narrow the pharynx. At the same
time, the larynx rises to shorten the distance between the vocal cords and
the narrowed zone (pharynx) and the widened zone (oral cavity). The
opposite occurs when the sound [i] is produced. In this case the tongue
rises, narrowing the oral cavity, and the pharynx widens. In the case of
[u], both the oral cavity and the pharynx widen, while the posterior part
of the tongue rises to narrow the posterior part of the palate.
For this complex articulatory coordination to take place, the oral cavity
and the pharynx need to be able to operate as two independent tracts, and
the pharynx therefore has to be of a certain length and also has to be
positioned at an angle to the oral cavity. In other words, the larynx needs
to be located low in the neck (Figure 16.1).
Vowel sounds formed in this way are modified in the oral cavity through
movements of the tongue, lips, and soft palate (where the uvula is
located), producing consonants. For the tongue to reach the correct
positions for production of consonants rapidly and precisely, it needs
not to be too long. This is because some of the muscles which move the
tongue insert into the hyoid bone, located in the lower and posterior part

Arsuaga / Chosen Species 1405115327-4-016 Final Proof page 247 8.6.2005 5:24pm

The Origins of Human Language

of the tongue (Figure 16.1). The longer the tongue, the longer these
muscles are, and the slower and less precise their movements.
It is easy to estimate the length of the tongue from a fossil skull, since it
is proportional to the length of the hard palate. However, deducing the
position of the larynx is a different matter. As we have already noted, the
larynx is formed of cartilage and supported by muscles and ligaments,
none of which fossilize.

The Fossils Speak

Since the mid-1970s, linguist Philip Lieberman and anatomist Jeffrey

Laitman have led a series of studies aimed at reconstructing the morph-
ology of the upper respiratory tract of fossil hominids. As a result of their
research they have concluded that a series of features in the base of the
skull can be used to determine the position of the larynx in the neck and
thus to establish the phonetic capacities of fossil hominids. The charac-
teristic most widely accepted by the scientific community as an indicator
of these capacities is the degree of flexion of the base of the skull.
If we were to cut through the human skull at its median line, or plane of
symmetry (the plane which divides the skull into two equivalent halves),
we would find that its lower edge shows a marked upward flexion between
the foramen magnum and the posterior part of the palate. However,
newborn humans and apes in general show very little skull base flexion.
In humans this flexion increases during childhood, reaching its maximum
in adulthood.
Since human newborns and apes both have a flat skull base, combined
with a high position of the larynx, and given that in humans the descent of
the larynx is accompanied by an increase in skull base flexion, there
appears to be a clear relationship between the position of the larynx and
the degree of flexion. This relationship has also been observed in experi-
ments on rats, in which an increase in skull base flexion was generated
On the basis of this premise, Laitman and his colleagues have carried out
various studies on different fossil hominids, and have drawn a number
of conclusions about their phonetic apparatus. According to these
researchers, in australopithecines, Paranthropus, and Homo habilis the
larynx must have been located high in the throat and their phonetic
capacities would have been similar to those of chimpanzees. However,
they found that the skulls from Broken Hill and Steinheim (from the

Arsuaga / Chosen Species 1405115327-4-016 Final Proof page 248 8.6.2005 5:24pm

The Origins of Human Language

Middle Pleistocene of Africa and Europe, respectively) showed flexion of

the skull base, implying a low position of the larynx and phonetic capacities
similar to our own. With regard to the Neanderthals, they came to the
conclusion that their phonetic apparatus allowed them to articulate only a
limited repertoire of vowel sounds (which did not include [a], [i], or [u]),
meaning that their spoken language would have been rudimentary and
However, we have studied the base of the skull in the only specimens of
Homo habilis (OH 24) and Homo ergaster (ER 3733) in which this part of
the skull is well preserved, and these show flexion values greater than
those of australopithecines, chimpanzees, and gorillas. These results sug-
gest that the phonetic apparatus of Homo habilis and Homo ergaster was
already similar to our own (albeit that in Homo habilis the palate was
proportionately as long as that of chimpanzees, indicating a more limited
repertoire of consonants). Thus they reinforce the hypothesis that all
species of our genus have had speech, given that, if these first humans
did not talk, the low position of their larynx is difficult to explain in terms
of natural selection.
Moreover, many researchers find it difficult to accept that the Nean-
derthals could have had a reduced capacity for speech compared to that of
their ancestors (for example, the Steinheim fossil).
In response to these criticisms, Laitman has suggested that the phonetic
capacities of Neanderthals were reduced as a result of an adaptation more
important for their survival, that of modifying their upper airways to
warm and humidify the cold, dry air of the ice ages: breathing is more
important than talking. Moreover, in their book In Search of the Nean-
derthals, Chris Stringer and Clive Gamble postulate that the ancestors of
the Neanderthals (Steinheim and Petralona) did not have a spoken lan-
guage like ours, despite having the anatomical structure for it, because of
the mental limitations of their relatively small brains.
However, the theory that Neanderthals were not able to speak as
we do began to look doubtful when, in the mid-1980s, paleoanthropolo-
gist Jean-Louis Haim announced that the skull of the Neanderthal
specimen known as the ‘‘Old Man,’’ from La Chapelle-aux-Saints,
had been poorly reconstructed by the first researchers and that his new
reconstruction showed a greater degree of skull base flexion. This
was confirmed by David Frayer, who measured the flexion in the new
reconstruction of the specimen from La Chapelle-aux-Saints and found
that it was similar to that of a series of medieval skulls. Given that this
fossil was one of those studied by Laitman, his results were thrown
into question.

Arsuaga / Chosen Species 1405115327-4-016 Final Proof page 249 8.6.2005 5:24pm

The Origins of Human Language

Moreover, in 1989 a hyoid bone belonging to a Neanderthal specimen,

the only one so far published for a fossil hominid, was found in the Israeli
deposit of Kebara. As we have already noted, the hyoid bone inserts into
the musculature of the tongue and its position in the throat is closely
related to that of the larynx. The Kebara hyoid bone shows a morphology
and dimensions comparable with those of the hyoid bone of any modern
human, leading the team of scientists who studied it, led by Baruch
Arensberg, to conclude that Neanderthals were anatomically just as cap-
able of speech as modern humans. This conclusion has been challenged by
Lieberman and Laitman, who assert that the morphology of the hyoid
bone is not relevant in determining the phonetic capacities of hominids.
Unfortunately, no skull has been found at Kebara, and it is thus not
possible to compare the morphology of the hyoid bone with the degree
of skull base flexion.
Research into reconstructing the phonetic apparatus of the Neander-
thals is thus currently at an impasse. Some researchers hold that the
studies of skull base flexion are not valid because they were carried out
on poorly reconstructed specimens, and they prefer to trust the results of
analysis of the Kebara hyoid bone. Others deny the validity of these
studies and continue to uphold the conclusions drawn from the analysis
of skull base flexion in the Neanderthals.
The only way out of this situation is to find new fossil material which
includes both intact skull bases and hyoid bones. It might seem that a find of
this nature is almost impossible, since it involves the highly unlikely combin-
ation of finding both an intact skull and a hyoid bone (only one is known in
the entire hominid fossil record, that of Kebara). However, such a discovery
has recently been made in La Sima de los Huesos deposit in Atapuerca, where
we found a skull with a virtually complete base, Skull 5 (see Figure 12.3), as
well as the majority of two hyoid bones. We will have to await the completion
of research currently under way on this extraordinary fossil material in order
to learn more about the origin of human speech.

Group Selection and the Extinction of the


In today’s world we are used to the idea that communication and infor-
mation are the key to progress, and the basis of our current technological
development. Thus it appears to us that possessing an articulated language
gives us an indisputable superiority over other living beings in the struggle

Arsuaga / Chosen Species 1405115327-4-016 Final Proof page 250 8.6.2005 5:24pm

The Origins of Human Language


Piriform sinuses

To the lungs

To the stomach

Figure 16.2 Section of the head of a nursing infant. The larynx is connected
to the nasal cavity and liquid passes into the alimentary canal through the piriform
sinuses. It is thus possible to drink and breathe at the same time. In adult humans,
the low position the larynx means that it has no a direct connection to the nasal
cavity, and breathing therefore has to be interrupted during the ingestion of

Arsuaga / Chosen Species 1405115327-4-016 Final Proof page 251 8.6.2005 5:24pm

The Origins of Human Language

for existence. But on reflection, it is easy to see that linguistic ability has no
advantage for a lone human confronting nature with only his own re-
sources: this characteristic only has meaning within a group to which the
individual necessarily belongs. Language is a property not of the individ-
ual, but of the collective. It is not the individual who communicates well;
it would be more appropriate to say that the group has good communi-
cation. The capacity to share and transmit information between individ-
uals and between generations by means of language confers a great
adaptive advantage on the group as a whole, but not on the isolated
This fact, which appears obvious to us, poses a serious problem for
evolutionary biology, because the theory of evolution through natural
selection – the Darwinian theory that is universally accepted, with minor
variations, by all scientists – is based on competition among individuals.
We have already noted that some authors have proposed a reductionist
version of natural selection, occurring at a level below that of the individ-
ual, such as the level of the genes. We have seen in this book that
competition can sometimes occur at the level of sperm – again, below
the level of the individual. Darwin himself put forward the mechanism of
natural selection in order to explain certain characteristics of individuals
which cannot be understood from the point of view of simple competition
for environmental resources and the struggle for survival; after all, what
matters is surviving in order to reproduce.
However, the appearance of properties which characterize groups, such
as communication, poses the problem of natural selection at the suprain-
dividual level, the level of the group. Thus, language was selected because
groups with a higher level of internal communication were more com-
petitive, more efficient in exploiting the resources of their environment,
and displaced other groups. In other words, an individual with the cap-
acity to produce and understand articulated language may be no more
competitive than another individual who does not have this ability, but a
group with language is more competitive than a group without.
It has been known since ancient times that competition between groups
of the same species occurs among social mammals, including primates.
Chimpanzees, to take a case in point, defend their territory and sometimes
invade that of other groups, or are invaded by neighboring groups. Not
only does communication serve to increase the cohesion and efficiency of
groups (‘‘information is power,’’ as we say in our society), but it appears
that behaviors of social cooperation (‘‘social altruism’’) within the group
are extremely important in selection between groups. Social cooperation
extends to many areas, such as defense of territory or resources, defense

Arsuaga / Chosen Species 1405115327-4-016 Final Proof page 252 8.6.2005 5:24pm

The Origins of Human Language

against predators, group hunting, caring for other adults’ offspring, shar-
ing food, and so on.
However, while competition between groups is easy to understand,
formulating a realistic model of group selection is a more complex matter.
If we start from the premise that cooperative behavior and the mental and
phonetic capacity for language have a genetic basis, we can approach the
problem by considering the existence of a gene for cooperative behavior
and a gene for language (this is, of course, an oversimplification, which we
hope the discriminating reader will forgive).
In order for selection between groups to occur, there must be a high level
of genetic homogeneity within each group – in other words, a high level of
consanguinity. To put it in more technical terms, the variability within each
group must be much lower than the variability between the different
groups. Groups in which the language and cooperation genes occur fre-
quently will be more competitive than others; since cooperation and com-
munication are not governed by the law of all or nothing, it would in fact be
the greater or lesser capacity for language and cooperation that made the
difference. In the case of social cooperation, there is the additional problem
that ‘‘selfish’’ individuals profit from the efforts of others without expend-
ing their own energies, and could thus benefit from natural selection and
leave more genes for the succeeding generation, with the result that ‘‘al-
truistic’’ behavior would never become dominant. One solution is that
there might be mechanisms of social rejection which prevent ‘‘selfish’’
individuals from reproducing, but then we are faced with the question of
how these mechanisms were selected, and we are back to square one.
Chimpanzees form social groups in which the males are related but not
the females, who move from one group to another. As we have seen, we
can picture something similar happening among the first hominids. Thus,
the social group is not a closed reproductive group, and hence genetic
uniformity is lost when females (genes, in short) are imported from other
groups. Even so, we can formulate mathematical models which make
selection at group level viable under certain conditions. But it is also
possible to imagine that our ancestors came to form larger social units,
which functioned as reproductive units and also competed with one
another. Whether or not this is the case, at present we have no better
mechanism than group selection to explain one of our most important
characteristics: articulated language.
Another question for which selection at group level appears the only
practical explanation is the replacement of the Neanderthals by modern
humans. It does not seem reasonable to suggest that this substitution took
place through selection at individual level. As we have already noted, the

Arsuaga / Chosen Species 1405115327-4-016 Final Proof page 253 8.6.2005 5:24pm

The Origins of Human Language

Neanderthals were physically stronger and their anatomy was better adapted
to the European climate (see Figure 13.1), and we therefore have to believe
that individually they were as well suited as modern humans to their envir-
onment (if not more so). It seems virtually certain that the success of our
ancestors was based on some group characteristic, but which one?
Many authors believe that the absence of a true language among
Neanderthals was one of the reasons, if not the main reason, for their
replacement by modern humans. To put it simply, the rudimentary lan-
guage of the Neanderthals limited their social complexity, restricting their
capacity to transmit information essential for exploiting the resources of
their environment. And when modern humans appeared on the horizon,
well equipped with their sophisticated language, the Neanderthals were
doomed to extinction.
Although the hypothesis of the linguistic superiority of modern humans
is an attractive one, we have already seen that it does not fit with the
fossils. Apart from the fact that it is not clear what kind of language
Neanderthals had, we have already noted that they were not ‘‘swept
away’’ by modern humans in a rapid, universal process: the replacement
of one by the other took place over about 10,000 years. If modern
humans had such an overwhelming advantage in terms of social complex-
ity and exploitation of the environment, why did it take them so long to
replace the Neanderthals?
We believe that what enabled our ancestors to displace the Neander-
thals was not the possession of a qualitative advantage, the type of lan-
guage, but rather the higher level of development of their capacities to
exploit resources: to put it simply, they had more of the same.

The Crooked Lines of Natural Selection

The men who discovered natural selection had an exemplary relationship.

There was never room for professional jealousy between Charles Darwin
and Alfred Russel Wallace; on the contrary, their relationship was one of
mutual respect and warm friendship. When, on April 26, 1882, British
society sought to recognize the importance of Darwin’s work by burying
him in Westminster Abbey, alongside the tomb of Sir Isaac Newton,
Wallace was among the three close friends who, together with members
of the aristocracy and the government, carried Darwin’s coffin.
Nevertheless, and despite their good friendship and mutual admiration,
Darwin and Wallace held opposing views on the origin of some of the

Arsuaga / Chosen Species 1405115327-4-016 Final Proof page 254 8.6.2005 5:24pm

The Origins of Human Language

most notable characteristics of our species, such as intelligence and

speech. While Darwin saw these traits as one further result of an evolu-
tionary process governed by natural selection, Wallace attributed them to
supernatural causes, placing the origin of human beings beyond the action
of natural selection.
One of the most profound arguments that Darwin advanced to support
the theory of evolution was the existence of ‘‘botched jobs’’ among living
beings. If organisms were the result of a direct act of divine creation, their
different parts should appear ‘‘as if just off the production line’’; in other
words, these parts should be specifically designed to perform a given
function efficiently. What we would not expect to find in this case is
organs that appear to be a more or less lucky modification of others
which perform a different function in other organisms. In other words,
we would expect each living being to have its own parts, perfectly tailored
to carrying out the functions with which they are entrusted.
Natural selection does not plan evolutionary change; it simply selects
from among what is there. That is to say, it preserves those variations in
existing organs which confer some advantage on individuals. Thus an
organ can be modified and end up performing a function different from
its original one. In this process, the organ in question may lose efficiency
in performing its original task, provided that this loss is compensated by
the advantage conferred by the new function.
But let us return to the theme of the human voice. When Wallace and
Darwin argued over the nature of natural selection and its role in the
origin of human beings, neither the anatomical basis nor the physiological
mechanisms of speech were known. Today we understand that this human
quality is based on the low position of our larynx, which in its turn is due
to a modification of the structure of the upper airways common in other
mammals. We also know that this particular position of the larynx restricts
our capacity to drink and breathe at the same time, and is responsible for
the unpleasant and dangerous phenomenon of choking. However, we
consider these to be minimal inconveniences compared with the great
advantage conferred by possessing a mechanism which allows us to modu-
late the sounds on which our language is based; there is no doubt that we
have made a very good bargain here. Thus, we can recognize the mark of
natural selection, and the trail of the evolutionary history of our species, in
the anatomy of our phonetic apparatus.
Darwin can rest easy at Newton’s side: once again, he was right.

Arsuaga / Chosen Species 1405115327-4-017 Final Proof page 255 4.6.2005 2:46pm


The Meaning of Evolution

Until tonight, you thought that life was absurd. From now on you will
know that it is mysterious.
Eric-Emmanuel Schmitt, The Visitor

The Action Replay of Life

In his book Wonderful Life, Stephen Jay Gould explains that we are not
the inevitable result of evolution, but rather a mere circumstance of it –
that if the videotape of life was rewound and played again from the
beginning, planet Earth would now be populated by a completely differ-
ent range of life forms, among which humans would be nowhere to be
It is, of course, impossible to perform this experiment by returning to
the beginning of life. However, to a certain extent it has been conducted
by natural means. For example, the platyrrhine monkeys in America did
not evolve into forms of intelligence comparable to our own. Clearly, they
were not subject to any ‘‘impulse’’ driving them toward ‘‘progress’’ or
‘‘perfection’’ (the same could be said of marsupials in Australia, and other
similar cases of evolution in conditions of geographical isolation).
In any case, there is another way we can play Gould’s game. If we
assume that evolution is directed or tends spontaneously toward increas-
ingly ‘‘higher’’ or more complex forms of life, we would expect the fossil
record to reflect a history in which progressively more complex forms of
life replace others, owing to their clear superiority, until humans appear.

Arsuaga / Chosen Species 1405115327-4-017 Final Proof page 256 4.6.2005 2:46pm

The Meaning of Evolution

But this analysis runs up against the problem of how to measure com-
plexity. We need to be able to establish a scale which could be applied to
living species or fossils, so that we can say: this species is at complexity level
3 or level 7 (the reader perhaps imagines that our species would be the only
one at level 10). We have to admit that this is a difficult problem to solve.
In 1949 George Gaylord Simpson, one of the greatest paleontologists
of the 20th century, published a very influential book entitled The Mean-
ing of Evolution, in which he concluded that evolution had no purpose.
Among other themes, in a chapter headed ‘‘Progress in Evolution’’
Simpson analyzed the question of whether there has been an increase in
complexity over the course of the Story of Life. It was clear to him that
when multicellular organisms (those with many cells) appeared, they
represented an increase in complexity over the first, unicellular (single-
celled) forms of life. According to Simpson, a second step toward greater
complexity occurred when the major different types of multicellular or-
ganisms emerged (the technical term for these types is phylum – plural
phyla); however, this progress took many directions, rather than following
a single, privileged line. From this point on, it is impossible to compare
the level of complexity within each of the lines. Simpson, who specialized
in vertebrate paleontology, wrote that it would be a brave man who would
attempt to prove that a human being is more complex than an ostraco-
derm (a kind of fishlike aquatic vertebrate which appeared more than
4 million years ago).
Meanwhile, we still have not defined what complexity is – no easy task
(you might also like to try). One modern way of doing it would be to use
the concept of complexity as it is applied to systems. A system is a com-
bination of elements which interact with one another, giving rise to the
properties of the system, and the more separate elements it has, the more
different possibilities for interaction exist, rendering the system richer in
functions, or more complex – in the sense of less predictable, less rigid,
more variable, and also more adaptable.
Multicellular organisms are self-regulated systems composed of differen-
tiated cells which form tissues and organs; these in their turn are organ-
ized into systems – respiratory, digestive, reproductive, excretory,
circulatory, or nervous. The concept of the complexity of systems can be
used to compare organisms in very different groups, for example to
compare mammals with sponges or jellyfish: the latter are clearly much
simpler forms of organization than mammals, with fewer differentiated
elements, and we can consider them biological systems with a relatively
low level of complexity. Who, however, would venture to assess the
relative complexity of a bat and a lion?

Arsuaga / Chosen Species 1405115327-4-017 Final Proof page 257 4.6.2005 2:46pm

The Meaning of Evolution

Even if we compare organisms in terms of the number of genes ex-

pressed (the number which translate into proteins), we find that protozoa
have more than bacteria, that invertebrates have more than protozoa, and
vertebrates more than invertebrates. However, among vertebrates it is
impossible to establish categories.
The issue of complexity is thus a real Gordian knot, and there is only one
way to undo it: by cutting right through it. We can start from the basis that
our species is, by definition, the most complex of all. However, if we
compare ourselves with other primates, or other mammals, we have to
ask what our greater complexity consists of. Only one of our systems, the
central nervous system, can be considered more complex, and it would have
to be this which ultimately won us top ranking on the scale of complexity.
Following this line of reasoning, which sets humans as the measure of all
things, it is clear that the closer a species is to Homo sapiens in its evolution,
the closer its appearance and relationship to us, the greater will be its degree
of complexity. Thus, mammals would be the most complex animals in
the history of life; the most complex mammals would be primates, and the
most complex of the primates, gorillas and chimpanzees, from which we are
differentiated by only about one percent of our genes.
For the time being we shall not challenge this dubious system of meas-
uring the degree of complexity of living beings using our species as a
model. Let it be so, if the reader so wishes. What certainly is debatable is
whether this supposed greater complexity constitutes an evolutionary
advantage leading toward the triumph (progressive, linear, and inexorable)
of the most complex. This would mean that the more complex would
always win out in the struggle for existence over simpler organisms – right
up to the arrival of the most complex being of all, the human being. Let us
look at what the fossil record says about this.
Mammals constitute the group of vertebrates to which we belong, and
this group is therefore universally considered the most ‘‘advanced’’ of all,
far ahead of amphibians, reptiles, or birds. We might therefore imagine
that since the time mammals appeared they have steadily replaced other
terrestrial vertebrates. Many people have the vague idea that mammals
emerged in a world dominated by dinosaurs, which they finally defeated
thanks to their superiority.
However, the fossil record indicates quite the contrary. The direct an-
cestors of mammals, a group of reptiles known as Therapsida (so similar to
the true mammals which came after that they have been called ‘‘mamma-
loid reptiles’’), dominated the terrestrial ecosystems at the start of the
Mesozoic (the Secondary era). This is normal, one might think – they
were superior to other reptiles. However, at a certain point, roughly

Arsuaga / Chosen Species 1405115327-4-017 Final Proof page 258 4.6.2005 2:46pm

The Meaning of Evolution

200 million years ago, Therapsida began to decline, and were gradually
replaced in terrestrial ecosystems by dinosaurs (birds later evolved from one
particular group of dinosaurs, making them the true ‘‘living dinosaurs’’).
If the term ‘‘superior’’ can be applied in evolutionary biology to any
particular group, in this case it would have to be applied to the dinosaurs.
The Therapsida eventually became extinct, and although some evolved into
the first mammals, these latter not only failed to win out over reptiles, but
led a very discreet life throughout the remainder of the Mesozoic (without
exception, all mammals were small in size). In fact it was the impact of a
meteorite, not the superiority of mammals, which finished off the dinosaurs
65 million years ago. If it had not been for this ‘‘providential’’ meteorite,
the evolution of terrestrial vertebrates would no doubt have been very
different. (Some authors believe that the extinction of the dinosaurs may
have been due to the atmospheric effects of a series of great volcanic
eruptions; the important thing for our argument is that the disappearance
of the dinosaurs had a nonbiological cause, and it is irrelevant whether this
was a meteorite, a volcanic phenomenon, or any other geological disaster.)
Let us now turn to another case from the pages of the fossil record, this
time even closer to ourselves. As we have noted in this book, within the
order of primates we belong to a group, the hominoids, which includes a
series of species, apes, with which we share many traits and, in truth, many
genes. In fact, apes are also the primates with the most developed brains.
Since they are the mammals most resembling us, it might be expected that
their superiority led them to win out at least over the other monkeys
existing at the time when they appeared. As we have noted, hominoids
originated in Africa at least 23 million years ago and, from the moment
when these primates made their way out of Africa to populate Europe and
Asia (about 17 million years ago), they became the group of primates with
the greatest evolutionary success, diversifying into a large number of
species inhabiting the wide belt of forest which extended throughout
much of the Old World. This might seem logical, since these were the
most intelligent primates and their success prefigured the glorious future
awaiting human beings.
Once more the fossil record tells us the opposite of what appears
‘‘logical.’’ Between about 8 and 7 million years ago, hominoids ceased
being the ‘‘kings of creation.’’ A great ecological change caused their
paradise to disappear. Astronomical factors, movements of continental
landmasses, and the emergence of mountain ranges changed the climate
and the composition of the atmosphere, resulting in deterioration of their
habitat over much of its once enormous expanse. The forest gave way to
more open ecosystems.

Arsuaga / Chosen Species 1405115327-4-017 Final Proof page 259 4.6.2005 2:46pm

The Meaning of Evolution

But it was not only that the change in vegetation reduced their living
space; other primates, the Cercopithecidae, became more numerous and
more varied than them. In military terms, the apes beat a retreat. Today
they remain reduced to two species of chimpanzee and the gorilla in
Africa, the orangutan in Sumatra and Borneo, and the gibbons of contin-
ental and island Southeast Asia. It is indeed significant that the little
gibbon, the hominoid least like humans, is the most successful in terms
of diversity and numbers. Despite pressure from humans, several million
gibbons, of nine different species, still survive.
But about 5 or 6 million years ago, in some place in Africa, possibly the
easternmost part, a specific type of hominoid began to emerge: the first
hominid, our oldest ancestor. To begin with it differed very little from the
ancestors of modern chimpanzees and gorillas. It could be considered the
East African version of the same group. Later, at least 4 million years ago,
this type of hominoid had developed a singular characteristic, never before
seen. This was not a hominoid more intelligent than the others. No – it
was a bipedal hominoid.
As time went on, bipedal hominoids adapted to the increasingly arid
ecosystems covering much of Africa. They developed specialized dentition
to cope with this. We have already seen that it is difficult to measure the
intelligence of fossil species (even of living species), but the index of
encephalization suggests that the most encephalized species 3 million
years ago were not hominids but dolphins.
By about 2.5 million years ago the hominids had split into two different
evolutionary lines. One of these, Paranthropus, developed a hypertrophied
(overdeveloped) masticatory apparatus. The other line was that of the first
representatives of the genus Homo, the first humans, who had a somewhat
larger brain. Only from this moment did hominids become unique among
living beings for their greater cerebral complexity. Paranthropus later became
extinct, and subsequent humans modified their body structure, increased
their brain size, and improved their technology. But even since the point
when intelligence appeared in the biosphere, human evolution has not
followed one single path, a straight line leading to ourselves. On the contrary,
until a few thousand years ago a number of intelligent human species existed
on earth. The fact that ours is the only one now in existence gives us the false
perspective that it has always been so, that our ancestors succeeded one
another in an ordered sequence, on a staircase we climbed step by step.
In short, neither the evolutionary history of mammals nor that of hom-
inoids reflects a pattern whereby these groups appeared and progressively
came to dominate other creatures thanks to their superior characteristics,
particularly their intelligence. On the contrary, the fossil record in both

Arsuaga / Chosen Species 1405115327-4-017 Final Proof page 260 4.6.2005 2:46pm

The Meaning of Evolution

cases shows a history of appearance and subsequent diversification, fol-

lowed by almost complete extinction and ultimate resurgence. In the case
of mammals, their resurgence was due to a favorable event of extraplanetary
origin (or some geological disaster). In the case of hominoids, the resur-
gence was only partial, and resulted from the adaptation of one of their
forms, the hominids, to a way of life completely new for primates, living in
open environments: cerebral complexity had nothing to do with this adap-
What does all of this mean? To put it simply, were it not for a series of
events unrelated to biology, such as a meteorite hitting the earth, the
formation of mountain ranges, great continental shifts, and other smaller
movements, we would not be here now to ponder these questions.
In other words, an extraterrestrial biologist at the beginning of the
Mesozoic might have predicted great evolutionary success for the ‘‘mam-
maloid reptiles’’ and their descendants, and would have been mistaken
(though incidentally, on this occasion the defeat of the mammaloid reptiles
occurred without the need for any great disaster: the dinosaurs ‘‘fought
clean’’ and defeated our ancestors purely through ecological competition).
Another alien biologist witnessing life on earth a few million years later
might have predicted a great future for the dinosaurs, and would also have
been mistaken. A third visitor, perhaps 10 million years ago, might have
said that hominoids would reign for ever in the forests of the Old World,
and would have been completely wrong.
The visitor from outer space who arrived 6 million years ago might be
convinced that the entire group of hominoids faced imminent extinction.
How could this extraterrestrial biologist have known that the ecological
change which had such a detrimental effect on hominoids overall would
favor the appearance of a type of bipedal hominoid which was later to give
rise to a species – our own – which would populate the world and end up
itself producing biologists? Even as little as 60,000 years ago, when the
Neanderthals had spread throughout Europe, Central Asia, and the Mid-
dle East, who could have predicted that modern humans, our ancestors,
would leave the African continent and be the cause of the extinction of the
Neanderthals a few thousand years later? And now that we are beginning
to know how things happened in the past, who would venture to make
predictions about the future of the biosphere?
But the really significant issue is not the possibility of imagining what
the future will be like – that is merely amusing speculation, a side issue.
The important thing is that our capacity to predict the future is the
measure of our knowledge of how the evolutionary process functions.
But does this knowledge really depend only on us? If evolution followed

Arsuaga / Chosen Species 1405115327-4-017 Final Proof page 261 4.6.2005 2:46pm

The Meaning of Evolution

specific tendencies or trajectories over time we could, by extrapolating

these into the future, predict its course. Since the only trend which
evolution appears to follow is that of adaptation in many different ways
to changing environmental circumstances, the question of where species
are going must remain unanswered.
This unpredictability of evolution indicates that nothing is decided
beforehand, that anything is possible. It shows that the most flourishing
biological group can become extinct because of changes in the physical
environment or through competition with other groups of organisms. No
form of life can be considered superior to others, because none is safe
from mass extinction.
But if evolution is unpredictable, does this mean that it is governed by
blind fate, that there are no laws, that everything is chaos and nothing can
be explained? Is it reasonable to accept that disorder (or a lack of order)
has produced such marvels of biology? Can random noise create a sym-
phony just by chance?
The very core of evolution is pure chaos. Natural selection operates on
genetic variants which arise without any relation to the activities or needs
of the organisms. Mutation, which generates variation, is a stochastic
process (regulated by chance). However, once a variant has arisen,
whether it survives and spreads, or is eliminated and disappears, does
not depend on chance. In the complex interrelation an organism main-
tains with other organisms and with the physical environment, certain
variants offer those who carry them a greater capacity for survival and
reproduction, while others reduce this capacity. Only the former will be
called to survive. Natural selection is a deterministic process.
But in the longer term, on the scale of species and groups rather than
that of individual organisms, does chance reign, or are there laws? Accord-
ing to traditional physics, including both Newtonian and the more mod-
ern quantum and relativist physics, total knowledge guarantees certainty,
and unpredictability is due purely to our lack of knowledge. However,
modern chaos theory predicts that there may be order, laws which we can
know, within a dynamic system, but that its future behavior can neverthe-
less still be unpredictable. How can we understand this apparent paradox?

Organization and Chaos

Today we fear that the agricultural and industrial activities of the human
race could destroy the ecological balance, causing the widespread

Arsuaga / Chosen Species 1405115327-4-017 Final Proof page 262 4.6.2005 2:46pm

The Meaning of Evolution

extinction of species, however well adapted they may be. The introduction
of animals, plants, or other organisms from one region into another (for
example, bringing rabbits to Australia) can also have catastrophic effects
on the ecological balance. We are likewise concerned about the effects on
the biosphere of climate change caused by the emission of greenhouse
gases such as carbon dioxide, and by the destruction of the ozone layer.
We are all increasingly aware that ecosystems, which are always in fragile
equilibrium, are composed of many species with a long history.
But all of these things about which we are (with reason) concerned have
occurred many times in the past, through natural causes. The continents
and oceans were not always distributed over the earth’s surface as they are
now, and the different regions where life is found have separated and come
together in many different ways, giving rise to various exchanges between
species in which some were favored while others lost out. Climate change
is a frequent occurrence in the Story of Life, and has a powerful impact on
ecosystems (the alternation between ice ages and interglacial periods such
as the one in which we are now living – which will not last forever – is but
one example). The composition of the atmosphere has also varied.
Finally, even if the physical environment remained unchanged, the
appearance, through evolution, of new species introduces a fundamental
instability factor into ecosystems, and means that they are always chan-
ging. Organisms show adaptations, developed over the period of their
evolution, which only make sense in relation to the ecological niches
occupied by their species. In a community, all populations are related to
one another in intricate networks, through which energy and matter flow.
To take the example of a primate, any change in the plants on which it
feeds or the animals it eats, the predators which prey on it, its competitors,
or its parasites, will have unpredictable consequences for the survival of
the species, which will have to evolve in its turn, adapting to its new
In other words, the biosphere is an enormously complex macrosystem,
made up of many different elements which organize themselves according
to a hierarchy of levels (cell, tissue, organ, system, individual, family
group, social group, population, community, ecosystem), and interact in
many different ways at all levels. This kind of system makes it extremely
difficult to identify the basic laws according to which it functions, even
when the constituent elements remain the same. However, to complicate
things further, the evolution of species means that the biosphere is an
unstable system, far from equilibrium, which has never remained static
and whose composition (species, and therefore their various interactions)
has changed over time.

Arsuaga / Chosen Species 1405115327-4-017 Final Proof page 263 4.6.2005 2:46pm

The Meaning of Evolution

It follows from all of this that the evolution of species is subject to the
influence of so many factors that in practice it is impossible to predict its
future. This does not mean that evolution depends on pure chance (in the
popular sense of incomprehensible chaos). On the contrary, it can be
understood, even if only with hindsight, like the weather. To a certain
extent, the well-known example of chaos theory called the ‘‘butterfly
effect’’ can be applied here: the flutter of a butterfly’s wings in Beijing
can cause rain to fall in New York (to say nothing of the case where the
delicate movement of an insect’s wings is replaced by a meteorite many
miles in diameter traveling at full speed toward earth).
But is this a merely technical question? Is our uncertainty due only to
the complexity of the problem? Chaos theory goes further, asserting that
even if we knew all these factors and interactions in full detail, the future
cannot be known, simply because it is not predetermined. This is the end
of certainty, which is replaced by probabilities.
In his book On Aggression, Konrad Lorenz recounts how Alfred Kühn
ended a lecture with Goethe’s words: ‘‘It is the greatest joy of the man of
thought to have explored the explorable and then calmly to revere the
unexplorable.’’ As the audience broke into applause, Kühn raised his voice
and cried: ‘‘No, not calmly, gentlemen; never calmly !’’ Should we assume
that at this point investigation into the nature of evolution is closed? Has
everything already been said?
We believe, on the contrary, that there is still much work to be done.
Newtonian physics speaks of trajectories which can be expressed in terms
of equations. If we know the initial conditions, these trajectories are
predictable and reversible, like a pendulum – first here, then there. In
these equations time does not exist: it is merely an illusion in which the
future and the past meet. Quantum physics simply replaces trajectories
with wave functions, but the symmetry with regard to time does not
change. Biological evolution, in contrast, is an irreversible process which
unfolds over time, surprising us at each moment, and follows no trajectory
(or trend). How can we reconcile physics and biology? If chaos theory is
correct, there is, as Ilya Prigogine (winner of the 1977 Nobel Prize for
Chemistry) says in his book The End of Certainty, a narrow path between
two equally alienating conceptions of the world. Either we live in a
deterministic world governed by immutable laws which leave no place
for novelty (and where the greatest novelty of all, evolution, would be
impossible), or we are in ‘‘a world . . . where everything is absurd, acausal
and incomprehensible,’’ subject to pure chance. It is the task of the
Darwins of the present and the future to travel this narrow path.

Arsuaga / Chosen Species 1405115327-4-epilogue Final Proof page 264 4.6.2005 2:46pm


Living organisms had existed on earth, without ever knowing why, for
over three thousand million years before the truth finally dawned on
one of them. His name was Charles Darwin.
Richard Dawkins, The Selfish Gene

The Never-ending Story

In his Dictionary of the Cinema, Fernando Trueba tells how the famous
French film director François Truffaut declared that when he went to the
cinema and saw a group of characters digging a tunnel for an hour and a
half and at the end of the film the tunnel was no use, he believed he should
get his money back. Although we cannot give our readers their money
back, we would not like to leave them with the unpleasant impression that
more than 3,000 million years of evolution have been completely point-
less, and that we are just a species like any other. Because in fact this is not
In his well-known book The Meaning of Evolution, which we referred to
in a previous chapter, George Gaylord Simpson explains that evolution has
no aim, but then rejects the suggestion that this means that man is ‘‘just
an animal.’’ To begin with, Simpson does not see any animal species as
‘‘just an animal,’’ because each one has its own peculiarities, shared with
no other species. But ours is exceptional in many very important aspects.
We are the most intelligent species, and this intelligence (however de-
fined) has enabled us to occupy all the different landscapes of the planet

Arsuaga / Chosen Species 1405115327-4-epilogue Final Proof page 265 4.6.2005 2:46pm


and to develop a technology. Whether this technology is used for good or

ill, for healing or for killing, we are capable of drastically altering our
environment and even the entire biosphere.
Eörs Szathmáry and John Maynard Smith point to a series of great
transitions in the Story of Life. Taking a look at these stellar moments of
evolution can help us to evaluate the true significance of our appearance in
the biosphere.
The first transition consisted of the transition from free molecules, or
‘‘replicants,’’ capable of replicating themselves, into populations of ‘‘repli-
cants’’ enclosed in a single container.
The second great transition consists of the association of ‘‘replicants’’ in
The third was the change in the composition of the inheritance mol-
ecule, which changes from being RNA (ribonucleic acid) to DNA (de-
oxyribonucleic acid). DNA contains the genetic information, while RNA
functions as an intermediary in the synthesis of proteins. With the appear-
ance of DNA comes the genetic code shared by all living beings (so in fact
these three steps are hypothetical, and actually form part of the Prehistory
of Life as we know it today).
The fourth transition moves from the organisms known as prokaryotes
(bacteria and blue-green algae or cyanobacteria) to the first eukaryotic
organisms. This is the category to which we belong, and it is characterized
by cells which have a nucleus and a series of organelles such as chloroplasts
and mitochondria. Many authors believe that the organelles are former
prokaryotes which have become integrated into the eukaryotic cell; mito-
chondria, for example, contain small circular molecules of DNA, like those
of bacteria.
The fifth great transition occurred when organisms moved from asexual
reproduction (‘‘self-cloning’’) to sexual reproduction.
The sixth transition, the next step, led from the organisms of the
kingdom of Protista (all unicellular and eukaryotic organisms) to multicel-
lular organisms, made up of many cells (these are also eukaryotes). This
step is thought to have occurred at least three times, independently, giving
rise to the animal and plant kingdoms and the kingdom of fungi. (In
addition to these four kingdoms there is also the Prokaryota kingdom,
mentioned above; some divide this in turn into two further kingdoms,
that of the ‘‘normal’’ bacteria, or Eubacteria, and a kingdom of a few
prokaryotes which live in extreme conditions and form the kingdom of
Archaebacteria). These six great transitions occurred at different mo-
ments, but all far back in time, since multicellular organisms have existed
for 680 million years.

Arsuaga / Chosen Species 1405115327-4-epilogue Final Proof page 266 4.6.2005 2:46pm


The seventh transition marked the shift from isolated organisms to

colonial organisms which include categories of non-reproducing organ-
isms, like the caste societies of insects.
Many of these transitions have some very interesting features in com-
mon. Essentially, we see a pattern where elements which live and repro-
duce in isolation lose some of their independence and come together to
form larger entities (and can no longer reproduce alone): the replicant
becomes integrated into the first container and then into the chromo-
some; the ancient, free-living prokaryotes become associated in the eu-
karyotic cell; the protist becomes a cell in a multicellular organism; the
individual has to live in a colony, in order to ensure the survival of genes
shared by the individuals; with sexual reproduction all organisms depend
on others to continue their line, and therefore need to belong to a
Moreover, several of these transitions are also marked by specialization
and division of labor among the elements which have come together:
the different genes code for different proteins; the cellular organelles
have different functions; the different cells of a multicellular organism
form tissues of very different kinds; each caste has its own function in the
Since we do not belong to a social insect species, we might take the view
that the most important events in the process of evolution which leads to
ourselves had already taken place at least 680 million years ago, and that
since then nothing of any major significance has happened. According to
Eörs Szathmáry and John Maynard Smith, nothing could be further from
the truth.
The eighth great transition took place a very short time ago, and
consisted in the shift from primate societies to human societies, with the
appearance of articulated language – a unique, revolutionary, and ex-
tremely powerful system for transmission of information (which has
made it possible, among other things, to write this book using an alphabet
of 26 letters). It has, after all, been worth the long wait. Although
genetically we are primates very close to chimpanzees, and a product of
evolution, we represent a type of organism radically different from all the
others. We are the only beings which question the meaning of our own
But let us not get carried away now by excessive triumphalism, because
we must also recognize that since the beginning of scientific thinking
among the ancient Greeks, there have been concerted efforts to position
our species with its back to nature, or even worse, above nature. This is the
source of some of the great problems afflicting humanity in the present.

Arsuaga / Chosen Species 1405115327-4-epilogue Final Proof page 267 4.6.2005 2:46pm


Only since Darwin has it been understood that we are not the chosen
species but, as Robert Foley suggests, a unique species among many other
unique species – albeit an amazingly intelligent one.
And it remains a paradox that so many centuries of science have led us
to know what any Kalahari Bushman, any Australian Aborigine, or any of
our ancestors who painted bison in the caves of Altamira knew full well:
that it is not the earth that belongs to man, but man who belongs to
the earth.

Arsuaga / Chosen Species 1405115327-5-bib Final Proof page 268 4.6.2005 2:47pm


In addition to the books cited in the text, we felt it would be useful to provide a
brief list of some of the most important recent publications which might be of
interest to the reader who wants to study human evolution in more depth.
Inevitably some important works will have been omitted, although we have
attempted to ensure that, though not all the references available are listed here,
all those that are listed are available.


Aguirre, E., ed.

1988 Paleontologı́a Humana. Barcelona: Prensa Cientı́fica.
Aiello, L., and C. Dean
1990 An Introduction to Human Evolutionary Anatomy. San Diego: Academic
Ayala, F. J.
1980 Origen y evolución del hombre. Madrid: Alianza.
Bermúdez de Castro, J. M., J. L. Arsuaga, and E. Carbonell, eds.
1992 Evolución humana en Europa y los yacimientos de la sierra de Atapuerca.
Valladolid: Junta de Castilla y León, Consejerı́a de Cultura y Turismo.
Bertranpetit, J., ed.
1993 Orı́genes del hombre moderno. Barcelona: Prensa Cientı́fica.
Bowler, P.
1986 Theories of Human Evolution. Oxford: Blackwell.
Brain, C.
1981 The Hunters or the Hunted? Chicago: University of Chicago Press.

Arsuaga / Chosen Species 1405115327-5-bib Final Proof page 269 4.6.2005 2:47pm


Conroy, G.
1990 Primate Evolution. New York: W. W. Norton.
Day, M.
1986 Guide to Fossil Man. London: Cassell.
Fleagle, J.
1988 Primate Adaptation and Evolution. San Diego: Academic Press.
Foley, Robert
1987 Another Unique Species. Harlow: Longman.
Fossey, Dian
1983 Gorillas in the Mist. Boston: Houghton Mifflin.
Goodall, Jane
1971 In the Shadow of Man. Boston: Houghton Mifflin.
1990 Through a Window. Boston: Houghton Mifflin.
Johansen, D., and M. Edey
1981 Lucy: The Beginnings of Humankind. New York: Simon and Schuster.
Johansen, D., and B. Edgar
1996 From Lucy to Language. New York: Simon and Schuster.
Jones, S., R. Martin, and D. Pilbeam, eds.
1992 The Cambridge Encyclopedia of Human Evolution. Cambridge:
Cambridge University Press.
Klein, R.
1989 The Human Career. Chicago: University of Chicago Press.
Landau, M.
1991 Narratives of Human Evolution. New Haven: Yale University Press.
Leakey, Richard
1981 The Making of Mankind. New York: E. P. Dutton.
Leakey, Richard, and R. Lewin
1992 Origins Reconsidered. New York: Doubleday.
Le Gros Clark, Wilfrid
1969 The Antecedents of Man. Chicago: Quadrangle.
Lewin, R.
1987 Bones of Contention: Controversies in the Search for Human Origins.
New York: Simon and Schuster.
Martin, R.
1990 Primate Origins and Evolution. London: Chapman and Hall.
Napier, J., and P. Napier
1985 The Natural History of the Primates. London: British Museum (Natural
Reader, J.
1981 Missing Links. Harmondsworth: Penguin.
Rightmire, G.
1990 The Evolution of Homo erectus. Cambridge: Cambridge University Press.

Arsuaga / Chosen Species 1405115327-5-bib Final Proof page 270 4.6.2005 2:47pm


Savage-Rumbaugh, S., and R. Lewin

1994 From Kanzi: The Ape at the Brink of the Human Mind. New York:
John Wiley.
Schultz, A.
1969 The Life of Primates. New York: Universe Books.
Stringer, Christopher, and Clive Gamble
1993 In Search of the Neanderthals: Solving the Puzzle of Human Origins.
London: Thames and Hudson.
Stringer, Christopher, and Robin McKie
1996 African Exodus. London: Jonathan Cape.
Szalay, F., and E. Delson
1979 Evolutionary History of the Primates. New York: Academic Press.
Tattersall, Ian
1995 The Fossil Trail. New York: Oxford University Press.
1995 The Last Neanderthal. New York: Macmillan.
Tattersall, Ian, E. Delson, and J. Van Couvering, eds.
1988 Encyclopedia of Human Evolution and Prehistory. New York: Garland.
Trinkaus, E., and P. Shipman
1992 The Neanderthals. New York: Vintage.
Walker, A., and Richard Leakey, eds.
1993 The Nariokotome Homo erectus Skeleton. Berlin: Springer-Verlag.
Walker, A., and P. Shipman
1996 The Wisdom of the Bones. New York: Alfred A. Knopf.
Wolpoff, M.
1980 Paleoanthropology. New York: Alfred A. Knopf.
Wood, B.
1991 Koobi Fora Research Project, vol. 4. Hominid Cranial Remains. Oxford:
Clarendon Press.


It is not our intention to give an exhaustive list of scientific articles, which

are generally published in specialist journals not easily accessible to the
majority of readers. However, we refer directly to some authors and papers
in our text, and it seemed right to us to cite the sources.

Aiello, L., and R. Dunbar

1992 Neocortex Size, Group Size, and the Evolution of Language. Current
Anthropology 34:184–193.
Aiello, L., and P. Wheeler
1994 Brains and Guts in Human and Primate Evolution: The Expensive Organ
Hypothesis. Current Anthropology 36:199–221.

Arsuaga / Chosen Species 1405115327-5-bib Final Proof page 271 4.6.2005 2:47pm


Arensburg, B., et al.

1990 A Reappraisal of the Anatomical Basis for Speech in Middle Palaeolithic
Hominids. American Journal of Physical Anthropology 83:137–146.
Arsuaga, J. L., et al.
1997 Size Variation in Middle Pleistocene Humans. Science 277:1086–1088.
Arsuaga, J. L., and I. Martı́nez
1989 Paleontologı́a humana: el origen de la humanidad. In Paleontologı́a.
E. Aguirre, ed. Madrid: CSIC.
Arsuaga, J. L., et al.
1993 Three New Human Skulls from the Sima de los Huesos Middle Pleisto-
cene Site in Sierra de Atapuerca, Spain. Nature 362:534–537.
Berge, C.
1900 Interprétation fonctionnelle des dimensions de la cavité pelvienne de
Australopithecus afarensis (AL 288–1). Zeitschrift für Morphologie und
Anthropologie 78:321–330.
Bermúdez de Castro, J. M., et al.
1997 A Hominid from the Lower Pleistocene of Atapuerca: Possible Ancestor
to Neanderthals and Modern Humans. Science 276:1392–1395.
Bromage, T., and M. Dean
1985 Reevaluation of the Age at Death of Plio-Pleistocene Fossil Hominids.
Nature 317:525–528.
Brunet, M., et al.
1995 Australopithecus bahrelghazali, une nouvelle espèce d’hominidé ancien de
la région de Koro Toro (Tchad). Comptes Rendus de l’Académie des Sciences
de Paris, série IIA 322:907–913.
Cabrera, V., and J. Bischoff
1989 The Dates for Upper Paleolithic (Basal Aurignacian) at El Castillo Cave
(Spain). Journal of Archaeological Science 16:577–584.
Cann, R., M. Stoneking, and A. Wilson
1987 Mitochondrial DNA and Human Evolution. Nature 325:31–36.
Carbonell, E., et al.
1996 Lower Pleistocene Fossil Hominids and Artefacts from Atapuerca-TD6
(Spain). Science 269:826–830.
Cavalli-Sforza, L.
1991 Genes, Peoples and Languages. Scientific American 265:72–79.
Cerling, T., et al.
1997 Global Vegetation Change through the Miocene/Pliocene Boundary.
Nature 389:153–158.
Coppens, Y.
1994 East Side Story: The Origin of Humankind. Scientific American 270:
deMenocal, P.
1995 Plio-Pleistocene African Climate. Science 270:53–58.

Arsuaga / Chosen Species 1405115327-5-bib Final Proof page 272 4.6.2005 2:47pm


Falk, D.
1993 Evolution of the Brain and Cognition in Hominids. James Arthur Lec-
tures on the Evolution of the Human Brain. New York: American Museum of
Natural History.
Fernández-Jalvo, Y., et al.
1996 Evidence of Early Cannibalism. Science 271:269–270.
Foley, R., and M. Lahr
1998 Mode 3 Technologies and the Evolution of Modern Humans. Cambridge
Archaeological Journal 7:3–36.
Gagneux, P., D. Woodruff, and C. Boesch
1997 Furtive Mating in Female Chimpanzees. Nature 387:358–359.
Hammer, M., and S. Zegura
1996 The Role of the Y chromosome in Human Evolutionary Studies. Evolu-
tionary Anthropology 5:116–134.
Harcourt, A.
1994 Sexual Selection and Sperm Competition in Primates: What are Male
Genitalia Good For? Evolutionary Anthropology 4:121–129.
Hublin, J. J., et al.
1995 The Mousterian Site of Zafarraya (Andalucia, Spain): Dating and Impli-
cations on the Palaeolithic Peopling Processes of Western Europe. Comptes
Rendus de l’Académie des Sciences de Paris, série II 321(10):931–937.
1996 A Late Neanderthal Associated with Upper Palaeolithic Artefacts. Nature
Johanson, D., et al.
1986 New partial skeleton of Homo habilis from Olduvai Gorge, Tanzania.
Nature 327:205–209.
Kappelman, J.
1997 They Might be Giants. Nature 387:126–127.
Kimbel, W., D. Johanson, and Y. Rak
1996 Systematic Assessment of a Maxilla of Homo from Hadar, Ethiopia.
American Journal of Physical Anthropology 103:235–262.
Klein, J., N. Takahata, and F. J. Ayala
1993 Mhc Polymorphism and Human Origins. Scientific American 269:78–83.
Krings, M., et al.
1997 Neanderthal DNA Sequences and the Origin of Modern Humans. Cell
Leakey, M. G., et al.
1995 New Four-million-year-old Hominid Species from Kanapoi and Allia Bay,
Kenya. Nature 376:565–571.
Lieberman, P., et al.
1992 The Anatomy, Physiology, Acoustics and Perception of Speech: Essential
Elements in the Analysis of the Evolution of Human Speech. Journal of Human
Evolution 23:447–467.

Arsuaga / Chosen Species 1405115327-5-bib Final Proof page 273 4.6.2005 2:47pm


Lovejoy, C.
1981 The Origin of Man. Science 211:341–350.
Manzi, G., A. Vienna, and G. Hauser
1996 Developmental Stress and Cranial Hypostasis by Epigenetic Trait Occur-
rence and Distribution: An Exploratory Study on the Italian Neanderthals.
Journal of Human Evolution 3:511–527.
McHenry, H.
1992 How Big were Early Hominids? Evolutionary Anthropology 1:15–20.
Moyà-Solà, S., and M. Köhler
1997 A Dryopithecus Skeleton and the Origins of Great-ape Locomotion.
Nature 379:156–159.
Parés, J. M., and A. Pérez-González
1995 Paleomagnetic Age for Hominid Fossils at Atapuerca Archaeological Site,
Spain. Science 269:830–832.
Plavcan, J., and C. Van Schaik
1996 Intrasexual Competition and Body Weight Dimorphism in Anthropoid
Primates. American Journal of Physical Anthropology 103:37–68.
Rosenberg, K., and Trevathan, W.
1997 Bipedalism and Human Birth: The Obstetrical Dilemma Revisited. Evo-
lutionary Anthropology 4:161–168.
Ruff, C., E. Trinkaus, and T. Holliday
1997 Body Mass and Encephalization in Pleistocene Homo. Nature 387:
Schrenk, F., et al.
1993 Oldest Homo and Pliocene Biogeography of the Malawi Rift. Nature
Semendeferi, K., et al.
1998 The Evolution of the Frontal Lobes: A Volumetric Analysis Based on
Three-dimensional Reconstructions of Magnetic Resonance Scans of Human
and Ape Brains. Journal of Human Evolution 32:375–388.
Seyfarth, R., and D. Cheney
1992 Meaning and Mind in Monkeys. Scientific American 267:122–129.
Suwa, G., et al.
1993 The First Skull of Australopithecus boisei. Nature 389:489–492.
Swisher III, C., et al.
1994 Age of the Earliest Known Hominids in Java, Indonesia. Science
1997 Latest Homo erectus of Java: Potential Contemporaneity with Homo
sapiens in Southeast Asia. Science 274:1870–1874.
Szathmáry, E., and J. Smith
1995 The Major Evolutionary Transitions. Nature 374:227–232.
Tague, R., and C. Lovejoy
1986 The Obstetric Pelvis of A.L. 288–1 (Lucy). Journal of Human Evolution

Arsuaga / Chosen Species 1405115327-5-bib Final Proof page 274 4.6.2005 2:47pm


Vrba, E.
1987 Late Pliocene Climatic Events and Hominid Evolution. In Evolutionary
History of the ‘‘Robust’’ Australopithecines. F. Grine, ed. New York: Aldine de
Wheeler, P.
1993 Human Ancestors Walked Tall, Stayed Cool. Natural History 102:65–67.
White, T., G. Suwa, and B. Asfaw
1993 Australopithecus ramidus, a New Species of Early Hominid from Aramis,
Ethiopia. Nature 371:306–312.

Arsuaga / Chosen Species 1405115327-6-ind Final Proof page 275 8.6.2005 5:21pm


Note: page numbers in italics refer to figures or illustrations

abductor muscles 66–7, 70–1 antelopes 53, 87

Aboriginal Australians 222–3, 267 Aotus trivirgatus 21
El Abric Romani deposits 214 apes 22; anthropoid 21, 66, 169–70;
Acheulian industry 107, 109, 185, 186, 206 birth 147, 150; diet 129–30, 131;
Adapiformes 28, 30 sexuality 155; teeth 135–6; see also
adaptation: evolution 261; fossils 58; primates
intelligence 169; Lamarck 14; natural Apidium 31
selection 10, 238; primates 19 Arago cave 191, 209
adaptive convergence 34, 110–11 Arambourg, Camille 89
Aegyptopithecus 31 Archaebacteria kingdom 265
Africa 22, 27, 226; fossil sites 52, 88, 99, Ardipithecus ramidus 51; forest-dwellers
174, 209; hominoids 32–3, 172; 142, 172; teeth 53, 112–13, 136–7, 138,
ocean influences 29, 43, 44; premodern 166
224–5; Proconsul 30, 30–2; L’Ardreda deposits 214
rainfall 43–4; see also individual Arensberg, Baruch 249
countries argon dating methods 60, 179
Africa, East 43–4, 87, 114–15, 186 arm–leg length 24, 33, 79, 104, 107
Africa, North 190 Arsuaga, Juan Luis 184, 187, 195
Afropithecus 30 art 212, 214–16, 218–19
agriculture 36, 130, 144, 261 Arvicola cantiana 185
Aguirre, Emiliano 184, 193 Asfaw, Berhane 51
Aiello, Leslie 127, 142–4, 169–70 Asia 172, 179–82, 190
alcelaphins 91 Atapuerca Mountains 191, 192, 196, 235,
alleles 50 249
Allen’s rule 200 Atlantic Ocean 29, 43, 44
allometry 118–19 Auel, Jean M. 216
altruism 164, 251–2 Aurignacian industry 212, 214
America, North 27, 28, 29, 37 Australia 190, 222–3
America, South 29 australopithecines 2, 58; bipedalism 73–4;
Amud site 221 body mass 122; brain size 81;
ancestors 15, 19–20, 22–4, 109–11, 216 dentition 86; encephalization 123;
Ankarapithecus 32 environment 162; facial skeleton 81, 86 ;
Annaud, Jean-Jacques 216 hands 101; ilium 81; life

Arsuaga / Chosen Species 1405115327-6-ind Final Proof page 276 8.6.2005 5:21pm


australopithecines (cont.) Border Cave 224

stages 153; monogamy 163; Boule, Pierre Marcelline 226
pelvis 150; sexual dimorphism 173; bovids 87, 91
vegetarian diet 142 Boxgrove site 182, 185, 191
Australopithecus afarensis: body mass 122; brachiation 23, 24, 25
brain size 123, 151–2; Brain, Bob 141
encephalization 173; Hadar area 54; brain 116, 117, 124, 125 ; asymmetry 126,
hands 141–2; jawbones 58–9; Laetoli 127; language 127, 241–3;
site 75–6; mastication 113; neck 73; lateralization 126, 127; lobes 124, 125,
phalanges 79; sagittal/nuchal 126, 127; mass 123, 128, 152, 169;
ridges 55, 81; sexual dimorphism 84, metabolism 142, 143
85, 165–6, 167, 168; skeleton 80, 82; brain size 15, 62, 117–20; anthropoids 21;
teeth 59, 137–9, 166 australopithecines 81; Australopithecus
Australopithecus africanus 61–2; brain afarensis 123, 151–2; birth
size 123, 151–2; disappearance 87; canal 151–2; body 118;
gender difference 63, 78; molars 146; chimpanzees 118, 119, 151–2; frontal
sagittal ridges 81; South Africa 173; lobe 127; Homo 102, 107, 113–14,
teeth 137–8, 139 143–4, 259; Homo ergaster 175;
Australopithecus anamensis 53, 54, 113, intelligence 127–8; longevity 153–4;
137–8, 172 Neanderthals 201; social behavior 169;
Australopithecus bahrelghazali 58, 59, 113 stomach size 142–4
Awash River site 51, 54, 60 Branisella boliviana 29
Ayala, Francisco 236 Bräuer, Gunter 226
breastfeeding 243, 250
baboons 18, 47, 135, 141, 161, 163, 166 Broca, Paul 128
Bañolas deposit 192 Broca’s area 117, 125–6, 241, 242
Bar-Yosef, Ofer 221 Broken Hill 209, 247
Baringo, Lake 98 Bromage, Tim 94, 98, 153
Barroso, Cecilio 215 Broom, Robert 87, 89, 91
bears 134, 186–7, 192 brow ridges: australopithecines 55, 81;
Berge, Christine 150 Gran Dolina 188; Hahnöfersand
Berger, Lee 141 fossil 215; Homo 102, 103, 105, 106,
Bergmann’s rule 200 113–14, 180; Neanderthals 203
Bering Strait 27 Brown, Peter 222
Bermúdez de Castro, José Marı́a 184, 187, Brunet, Michel 58
194–5 Brunhes chron 185, 191
Biache-Saint-Vaast deposit 191, 208 burins 212, 213
Bilzingsleben deposit 191, 207, 209, 218 burying the dead 206–7, 218
biomechanical efficiency 77 Byrne, Richard 170
biosphere 262
bipedalism 3, 9, 62, 259; abductor Cabezo Gordo deposit 192
muscles 70–1; calcium 205
australopithecines 73–4, 113; Callitrichinae 21
ecological niche 169; Can Llobateres deposit 32
food-carrying 164; hominids 53, 54; Canidae 134
legs 24; Lucy 1–2, 71, 76; canines 132, 134–6, 137, 139, 166
monogamy 163, 166, 167, 168 ; Cann, Rebecca 230, 237
pelvis 24; savanna 76; skeleton 77–8; cannibalism 64, 187, 192
tools 65, 77 carbohydrates 130, 134
birth canal 147, 148, 149, 150–2, 200 carbon-14 dating method 60
birth intervals 164 carbon dioxide 40, 44–8, 172, 262
Bischoff, James 194, 214 Carbonell, Eudald 184, 186, 187
Black Eve hypothesis 230–1 Caribbean Sea 29
Bolomor molar 191 carnivores 129–30, 132, 134, 142–4,
Bonis, Louis de 32 186–7
bonobos 20, 22, 24, 101, 160 Carretero, José Miguel 195

Arsuaga / Chosen Species 1405115327-6-ind Final Proof page 277 8.6.2005 5:21pm


carrion 143, 186–7 chromosomes 228–9, 231–2, 236, 265

El Castillo cave 214 chron 61, 185
catarrhines 21–2, 111, 145–6 clades 22
catastrophe 12, 39 Clark, Sir Wilfrid Le Gros 65, 141, 152–3
Cavalli-Sforza, Luigi 232 Clarke, Arthur C. 49, 64
cave bear 134, 192 Clarke, Ron 141
cave-dwellers 186–7 clavicle 24, 111
cave paintings 214–16, 219 cleavers 107, 109
cellulose 130, 134 climate change: Africa, East 87; average
Cenozoic 27 temperature 37 ; carbon dioxide 262;
Ceprano site 191 catastrophe 39; global 172;
Cercopithecidae 22, 47, 162, 172, 259 grasslands 143; habitats 87;
Cercopithecus aethiops 240, 241 intelligence 64–5; latitude 37, 44;
cereals 130, 139, 144 Paranthropus 175
cerebellum 116, 117 colobus monkey 25, 51, 53, 135
cerebral cortex 117, 124, 241, 242–3 comparative sociobiology 158–60
cerebrum 116, 117, 124 competition 47, 144, 251
Cerling, Thure 46 complexity 256–7
cetaceans 120 Conrad, Joseph 145
Chad fossils 58–9 continental drift 40, 44
chaffinches 35–6 Coon, Carleton 227
La Chaise-Abri Suard deposit 191, 208 cooperation 251–2
chaos theory 261, 263 Coppens, Yves 56, 57, 59, 89
La Chapelle-aux-Saints 248 copulation 150
Chatelperronian industry 213–14, 216, Cosquer, Grotte 215
217 La Cotte de St. Brelade 207
Chauvet, Grotte 214–16 coxal bones 69–70, 71, 191
Chemeron site 51 cranial capacity 125, 196
Cheney, Dorothy 240 Cretaceous 27
Chenjiawo site 181 Crichton, Michael 233, 235
childbirth 70, 146–7, 148, 149, 150 Cro-Magnon man 134, 215, 216–19
children 62, 64, 197, 207 culture 158
Chilo 171 Curtiss, Garniss 179
chimpanzees, common (Pan
troglodytes) 22, 23, 49–50; airways 244 ; Dar-es-Soltan II cave 224
birth 147, 148; brain size 118, 119, 123, Dart, Raymond 61–2, 64, 141
151–2; diet 130; distribution 162–3, Darwin, Charles 264, 267;
217, 259; estrus 146, 155–6; extensor Autobiography 9; Beagle trip 35–6; The
muscles 69 ; femur 73; forest habitat 79; Descent of Man 1, 197; evolution of
gluteus medius 67 ; hamstring species 3, 14, 110; natural
muscles 72 ; hands 101; selection 10–13, 254; On the Origin of
language 240–1; life expectancy 146; Species 197; sexual selection 160;
locomotion 26–7, 77–8; Wallace 10, 253–4
menstruation 165; pelvis 68; sexual Darwinism 16, 171
dimorphism 159–60; skeleton 25 ; Dawkins, Richard 164, 264
social groups 159, 170, 251–2; days/nights 41
teeth 138–9, 146; testicles 160–1; Dean, Chris 153
tools 98–100 Dederiyeh site 221
chimpanzees, pygmy (Pan paniscus) 20, 22, deMenocal, Peter 44
160, 217, 259 dentition: australopithecines 59, 86,
China 102, 181, 185, 209, 224 172; carnivores 134; catarrhines
Chinese medicine 34 21–2; hominoids 32–4; life
chloroplasts 265 stages 145–6; Oreopithecus 34;
choking 243–4, 254 Paranthropus 34, 59, 95, 143;
chopping tools 100 primates 19–20; see also teeth
chordates 15 Descartes, René 116

Arsuaga / Chosen Species 1405115327-6-ind Final Proof page 278 8.6.2005 5:21pm


diastema 135, 139 evolution 39–40, 251, 254, 260–1, 263

diet 129–30, 132, 139, 141, 169, 172; see evolution of species: Darwin 3, 14, 110;
also carnivores; herbivores; vegetarians directionality 14; divergence 15;
digestion 134, 142–3 Lamarck 9–10; natural
Dinesen, Isak 226 selection 10–13; progress
dinosaurs 39, 258, 260 concept 13–16; Story of Life 15, 109,
Dmanisi site 182 256, 262, 265
DNA (deoxyribonucleic acid): evolutionary diagrams 29, 112, 114, 189,
fossils 233–4; humans 234–5; 210
inheritance 228–9, 265; extension 69, 71
mitochondrial 229–31, 232, 233–4;
nuclear 231, 232 facial characteristics: australopithecines 81,
dolphins 110, 111, 120, 169, 259 86 ; Gran Dolina Child 188; nasal
drupes 131 bones 105, 106, 188; Neanderthals 193,
Dryopithecus 32, 33 202, 203, 204, 205, 208–9; Sangiran
Dryopithecus laietanus 33 site 181; see also brow ridges; nuchal
Dubois, Eugène 179, 180 ridge
Dunbar, Robin 127, 169–70 Falguères, Christophe 194, 223
Falk, Dean 125–6, 127
earth–sun orbits 40–3 El Fayum deposits 28, 31
East Side Story hypothesis 44, 57–9 Felidae 17, 132
echinoderms 15 femur 73, 79, 104, 107
ecological factors 35–6, 46–8, 261–2 La Ferassie 1 site 198
ecological niche 10, 76, 94, 160, 169, figurines 215, 218
172 fire 206–7
education 158 first family concept 56, 78
Egypt 28, 31 ; see also El Fayum fission-track dating 60
Ehringsdorf deposit 191, 208 Foley, Rob 162, 166, 209, 232, 267
Edelweiss group 192–3 footprints, Laetoli site 74–6
Einstein, Albert 5 foramen magnum 73, 74, 81, 247
Eldredge, Niles 12 foraminifera 38–9
electron spin resonance 60–1 forest habitat: chimpanzees 79; gorillas 79,
elephant 46, 118 168 ; hominids 51, 53; Miocene 57;
enamel on teeth 53, 113, 136–7, 153 primates 19; reduction 46–7, 48, 258
encephalization 119, 120; Fossey, Dian 158
australopithecines 123; Australopithecus fossils: dating 59–61; DNA 233–4;
afarensis 173; hominids 117, 123–4, hominids 51, 53, 57–9, 172–3; pollen
259; humans 128; Neanderthals 201 grains 39; sedimentary deposits 58, 59;
end-scrapers 212, 213 sex of 195–6; see also paleontology
energy 132, 169 Frayer, David 248
Engis site 197 Frisch, Karl von 156
environmental factors 11–12, 162 frontal lobe 124, 125, 126, 127
Eocene 28 fruit 130, 131, 159
epiglottis 244 Fuente Nueva-3 186
Equidae 46, 91 funerary practice 194, 206–7, 218
equinox 40, 41, 42
Erythrocebus patas 18, 47 Gagneux, Paul 159
estrus 146, 155–6, 159, 165 Galápagos Islands 35–6
Ethiopia 51 Galdikas, Biruté 158
ethology 156, 157 Gamble, Clive 207–8, 248
Eubacteria kingdom 265 Garrod, Dorothy 221
eukaryotes 265 geese 156
euprimates 27, 29 geladas (Theropithecus gelada) 18, 47, 135,
Eurasia 27, 28, 34, 37 141, 163, 166
Europe 27, 172, 182–6, 191, 211; see also gender difference 78, 164–5; see also sexual
individual countries dimorphism

Arsuaga / Chosen Species 1405115327-6-ind Final Proof page 279 8.6.2005 5:21pm


genes 228; behavior 157; Holliday, Trenton 198

chromosomes 236; difference 49–50, Holloway, Ralph 125–6
216–17; flow 227; natural hominids 15, 59, 76, 112–15;
selection 156, 164, 261; neutral/ bipedalism 53, 54;
non-neutral 50 encephalization 123–4; fossils 51, 53,
genetic defects 238 57–9, 172–3; hunting 64;
genetic manipulation 238 mandible 54; phonetic
geochronology 61, 179, 194 apparatus 247–8; prey/predators 141;
Gibart, Josep 182 teeth 136–8
gibbons (Hylobates lar) 22, 23, 24, 161; hominoids 22–7, 33–4, 111; Africa 30–4,
locomotion 26; monogamy 163; 172; comparative
species 217, 259 sociobiology 158–60; film 64;
Gibraltar, Straits of 189, 190 Miocene 30, 47; species 258
Gibraltar site 197 Homo spp. 99 ; brain size 102, 107,
Gigantopithecus 32 113–14, 143–4, 259; emergence 89,
Gigantopithecus blacki 34 142; evolutionary diagrams 29, 112, 114,
glottis 244 189, 210 ; jawbones 94, 98; sexual
gluteus muscles 67, 68–9, 70, 71 dimorphism 102; skulls 102
Goldschmidt, Richard 12 Homo antecessor 187–90, 210, 218, 224, 235
Gongwangling site 181 Homo erectus 102, 179–82, 190, 222–3
Goodall, Jane 158, 162, 170 Homo ergaster 102, 108, 114, 181; body
gorillas (Gorilla gorilla) 22, 23 ; birth 147; mass 122, 152; brain 123, 127, 152,
brain size 118; diet 130; estrus 155–6; 154, 169, 175; brow ridge 105, 106 ;
foramen magnum 74 ; forest habitat 79, megadonty 139; skull 248; social
168, 259; language 240; change 169; Swartkrans 107
locomotion 26–7, 77–8; muscles 72, 81; Homo habilis 102, 114; arm–leg
sexual dimorphism 159; length 107; body mass 122; brain 123,
sociability 158–9, 165; teeth 135, 138; 127, 169; brow ridge 103 ;
testicles 160–1 megadonty 139; Olduvai Gorge 104;
Gould, Stephen Jay 12, 255 parietal lobe 242; skull 248
Gracia, Ana 195 Homo heidelbergensis 209, 210, 217, 218,
Gran Dolina 184–6, 187–90, 192, 210 224
grasslands 48, 134, 143, 173 Homo neanderthalensis 216–19
Great Rift Valley 57, 173 Homo rhodesiensis 224
greenhouse effect 40, 45–6, 262 Homo rudolfensis 98, 102, 104, 114, 123,
Greenland 37, 40 127
gyri 124, 125, 242 Homo sapiens 13, 97–8; African
origins 223–5; dentition 21–2;
habitat changes 58, 64–5, 87, 172 encephalization 120; Mode 3
Hadar site 54, 56–7, 76, 98 hypothesis 209; and
Haeckel, Ernst 180 Neanderthals 216–19; see also humans,
Hahnöfersand fossil 215 modern
Haim, Jean-Louis 248 Howell, Francis Clark 219
hamadryas baboons 163, 166 Hublin Jean-Jacques 214, 215
Hammer, Michael 231 humans, modern 23, 25 ; airways 245 ; birth
hamstring muscles 71, 72 canal 147, 148, 149, 150; brain
handaxes 107, 109 size 117–20, 128; and chimpanzees 50;
hands 24, 101, 141–2 diet 129–30, 131; DNA 234–5;
haplorrhine primates 20–2, 28, 111 encephalization 128; extensor
haplotypes 231 muscles 69 ; foramen magnum 74 ;
Harcourt, Alexander 160–1 gluteus medius 67 ; hamstring
Henry VIII, King of England 37 muscles 72 ; jaw/teeth 19 ;
herbivores 45–6, 134, 142–3, 186–7 menstruation 146, 165; nature 266–7;
heredity 10; see also inheritance Neanderthals 212–16; pelvis 70;
hierarchies 13–14, 22–3 sexuality 155–6; skulls 202 ; spine 25,
histocompatibility complex 236 73, 79; see also Homo sapiens

Arsuaga / Chosen Species 1405115327-6-ind Final Proof page 280 8.6.2005 5:21pm


humerus 24, 79, 104, 107, 191 Krakatoa 39

hunter-gatherers 131, 164 Krapina fossils 211
hunting 64, 101–2, 143 Krings, Matthias 233
hydrosphere-atmosphere system 40 Kromdraai deposit 89, 91
hyenas 132, 187 Kubrick, Stanley 64
Hylobates lar: see gibbons Kühn, Alfred 263
hyoid bone 245, 246, 249
Laetoli site 54, 56–7, 74–6
Ice Ages 36, 37, 44 Lahr, Marta 209
Ice Man of Tirol 233 Laitman, Jeffrey 247, 248, 249
iliac crest 69–71, 73, 146–7, 148 Lamalunga cave 191
ilium 69, 71, 81 Lamarck, Jean-Baptiste de 9–10, 14, 66,
incisors 134–6, 153–4 129, 157
India 224 Lana (chimpanzee) 241
Indian Ocean 44 Landau, Misia 2
infants 11, 151–2, 165, 200 language: articulated 62, 64, 266;
information technology 157 brain 127, 241–3; chimpanzees 240–1;
inheritance 111, 228–9, 265 communication 249, 251;
intelligence 116, 259–60; adaptation 169; Lorenz 239–40; Neanderthals 253
carnivores 143–4; climate 64–5; laryngeal tone 244–6
Neanderthals 217–18; physical larynx 243–4, 245, 246, 250, 254
features 124, 127–8, 237–8; Leakey, Louis 89, 104, 181
primates 15; technology 264–5 Leakey, Mary 53, 74, 89, 104
Inuit 131, 200 Leakey, Richard 53, 89, 91, 104, 105
ischium 69, 71, 150 legumes 130, 139, 144
isolation 238, 255 lemurs 20, 21, 28, 119
isotopes 38, 190–1 Levallois technique 205–6
Israel 185 Levant 217, 219, 221–2
Lezetxiki humerus 191
Java 102, 179–82, 185, 223 Lieberman, Philip 247, 249
jawbones: adolescent 188; Australopithecus life expectancy 146, 153–4
afarensis 58–9; hominoids 34; life stages 145–6, 152–4
Homo 94; Paranthropus 96, 113, 121; Linné, Karl von 22–3, 97
teeth 133 ; see also mandibles; maxilla lions 17, 133
Jebel Qafzeh deposit 219, 220, 223–4 Lockwood, Charles 166
jerboas 53 locomotion 26–7, 64; see also bipedalism
Johanson, Donald 2, 54, 56, 78, 98, 104 Longgupo deposit 181
Jungers, William 76, 122, 166 Lorenz, Konrad 156, 239–40, 263
Lorenzo, Carlos 195
Kanapoi site 137 Lorisidae 21, 28
Kanzi (bonobo) 101, 240–1 Lothagam deposit 51
Kappelman, John 198 Lovejoy, Owen 150, 163–5, 166, 167,
Kebara 221, 249 168
Keith, Arthur 23–4, 142, 219 Lucy: age of fossils 56; bipedalism 1–2, 71,
Kenya 48, 51 76; discovery 54; pelvis 150;
Kenyapithecus 30 proportions 78, 79; pubis 151
Kidd, Judith and Kenneth 232 Lufengpithecus 32
Kimbel, William 166
Kipling, Rudyard 197 macaques (Macaca sylvannus) 18, 161
Klasies River Mount 224 McCown, Theodore 219
Koenigswald, Ralph von 34 Machado, Antonio 97
Köhler, Meike 32 McHenry, Henry 122, 138–9, 166
Konso deposit 91, 105 Machiavelli, Niccolò 170
Koobi Fora 102, 104, 105, 107 Madagascar 21, 29–30, 119
Koufos, George 32 magnetic field changes 61, 185, 191
Krainitzki, Heike 233 Makapansgat site 62

Arsuaga / Chosen Species 1405115327-6-ind Final Proof page 281 8.6.2005 5:21pm


Malawi, Lake 94 Moyà-Solà, Salvador 32

Malthus, Thomas Robert 10–11 multicellular organisms 256–7, 265,
mammals 14, 87, 118–19 266
mammoths 233 multiregional origin hypothesis 227–8
mandibles: Africa, North 190; Bañolas Mustelidae 134, 192
deposit 192; hominids 54; Homo 98; mutation 12, 50, 261
Kanapoi 137; Lothagam 51;
Mauer 182, 185, 187, 191, 209–10; nails, flat 19, 111
Paranthropus 94–5, 96 Napier, John 104
mandibular ramus 96 nasal bones 105, 106, 188
mandrills 135 nasal cavity 243–4
Manzi, Giorgio 201, 225 natural selection: adaptation 10, 238;
marine paleotemperature scales 38–9, 43, Darwin 10–13, 160, 254;
190–1 evolution 251, 254; genes 156, 164,
Martı́nez, Ignacio 187, 195 261
mass spectrometer 60 Nature 136, 230
masseter muscles 93, 96 Neanderthals 169, 191, 211; brain
mastication 93, 94–5, 113, 173, 259; see size 201; brow ridge 203; burying the
also dentition; teeth dead 206–7, 218; Cro-Magnon
Matuyama chron 185, 191 man 216–19; encephalization 201; facial
Mauer mandible 182, 185, 187, 191, characteristics 193, 202, 203, 204, 205,
209–10 208–9; fire 206–7; humans 212–16;
maxilla 93, 95, 96, 137, 188, 203 intelligence 217–18; language 253;
Mayor Cave 192, 194 mitochondrial DNA 233–4;
megadonty, index of 138–9 neurocranium 201, 202 ; phonetic
meiosis 228 apparatus 247; physical
menarche 146 characteristics 198, 199, 200, 253;
Mendel, Gregor 10, 226 skulls 201, 202 ; Spain 211; see also
meninges 125 Mousterian industry
menstruation 146, 165; see also estrus neo-Darwinism 12
Mesozoic 257–8, 260 neocortex 124, 127, 170
metabolism 142, 143 Neolithic 130
meteorite impacts 39, 258 neurocranium 120–1, 179–80, 201,
Michelangelo 100 202
Milankovitch Cycles 40–3 newborns: see infants
Mimomys savini 185 Newton, Sir Isaac 253
Miocene 30, 47, 57 Ngandong remains 180, 223
missing links 89 North Pole 40
mitochondria 265 nose 20–1, 86; see also nasal bones
mitochondrial DNA 229–31, 232, 233–4 Notarctus 30
Mladec deposit 215 notoungulates 46
Modjokerto child 179 nuchal plane 73, 81
molars 135–7, 139, 191; nuchal ridge 55, 81
catarrhines 145–6; chimpanzees 146; nuts 131, 141, 144
gorillas 135; life stages 154;
Paranthropus 153; size 138–41 occipital lobe 124, 125
molecular biology 49 occipital ridge 180
molecular clocks 49–50, 237 Old World monkeys 22, 47, 135, 162,
monogamy 161–3, 165, 167, 168 172
Montaña de Monserrate, Bernardino 239 Oldowan industry 100–2, 107
Morotopithecus 30 Olduvai Gorge 104, 107, 181
Mosquero, Marina 187 Oligocene 28
mouse lemur 119 Olorgesailie deposit 48
Le Mouster 207–8 Omo River 89, 91, 97, 104, 224
Mousterian industry 205–6, 212, 214–15, Omomyidae 28
217, 221 oral cavity 243–4, 244

Arsuaga / Chosen Species 1405115327-6-ind Final Proof page 282 8.6.2005 5:21pm


orangutans (Pongo pygmaeus) 22, 23, 34, photosynthesis 14, 45–6

259; birth 147; diet 130; phylogenetic relationships 110, 111
females 155–6; hominoid fossils 33; phylum 256
humans 50; locomotion 26; social physics 263
life 158, 165; temporal muscles 81; Pinilla del Valle deposit 191
testicles 161 Pithecanthropus 179
Oreopithecus 34 Pithecanthropus erectus 180
organelles 229, 265 plant types 45–6
Ouranopithecus 32, 33 platyrrhine primates 21, 29, 30, 255
Out of Africa hypothesis 226, 236 Plavcan, Michael 166
owl monkey (Aotus trivirgatus) 21 Pleistocene 47–8, 185; Middle 191, 207;
oxygen 38–9, 190–1 Upper 211
ozone layer 262 Plesiadapiformes 27, 28
Pliocene 47
Pääbo, Svante 233 pollen grains 39, 207
paleoanthropology 4–5, 51, 53, 56, 227 polymorphism 230, 231, 232
paleocortex 124 Pongo pygmaeus: see orangutans
paleomagnetism 61 porphyria 236
paleomorphology, functional 58, 68 positron emission tomography 242
paleoneurology 125, 241–3 precipitation 40
paleontology 4, 56, 108–10, 121 premolars 134, 135, 136, 138
Pan paniscus: see chimpanzees, pygmy Prigogine, Ilya 263
Pan troglodytes: see chimpanzees, common primates 17–22; catarrhine 21–2, 111,
Papua New Guinea 190 145–6; evolutionary relationships 29 ;
Paranthropus 88, 114–15; body mass 122; fossils 16, 27; haplorrhine 20–2, 28,
brain size 143; climate change 175; 111; history of 27–34; intelligence 15;
competition 144; dentition 34, 59, 95, platyrrhine 21, 29, 255;
143; diet 139; emergence 87, 89, 142; skeletons 25 ; strepsirrhine 20–2;
environment 162; jawbones 93, 95, 96, teeth 134–6
113, 121; life stages 153; Proboscidea 46
mastication 94–5, 173, 259; sagittal Proconsul 30, 30–2
ridge 90, 91, 92, 95; sexual prognathism 86, 92, 93, 113, 203
dimorphism 173; teeth 139, 140, 153 prokaryotes 265, 266
Paranthropus aethiopicus 89, 90, 173 promiscuity 159
Paranthropus boisei 47, 89, 173; brain protein 130, 131, 132, 139, 143
mass 123; skulls 91, 92 ; teeth size 139, Protista kingdom 265
140 Proverbs, Book of 35
Paranthropus robustus 89, 115, 173; brain puberty 145, 164
mass 123; Swartkrans 91, 107, 141; pubic bone 200
teeth 139 pubis 70, 150, 151
Parés, Josep Marı́a 185 Purgatorius ceratops 27
parietal lobe 124, 125, 242 Pusey, Anne 170
patas monkey (Erythrocebus patas) 18, 47
paternity 165 races 227
patrilocal societies 232 radioactive decay 60
pelvis 66–7, 69–71; rainfall 40, 43
australopithecines 150; bipedalism 24; rainforests 18, 172
childbirth 70, 146–7, 148, 149 ; Rak, Yoel 54
chimpanzees 68; hyperfeminine 151; La Ramón y Cajal, Santiago ix
Sima de los Huesos 201; Turkana recombination 228–9
Boy 152 Reilingen deposit 191, 208
Pérez-González, Alfredo 185 Renne, Grotte de 214, 216
Petralona deposit 191, 208, 209, 248 replicants 265
phalanges 79, 182 reptiles 257–8, 260
pharynx 243–4, 246 resonance 246
phonetic apparatus 247–8 Retzius, striae of 153

Arsuaga / Chosen Species 1405115327-6-ind Final Proof page 283 8.6.2005 5:21pm


rib cages 25, 79, 82 Sivapithecus 32–4, 50

Richmond, Brian 166 Skhul rock shelter 219, 221, 223–4
Robinson, John 91 skulls 81, 209; Homo 102, 248;
rodents 30, 185 Neanderthals 197, 201, 202 ;
Rosas, Antonio 187 neurocranium 120–1; Saccopastore 211;
Rosenberg, Karen 147, 150 Sangiran 179–80; La Sima de los
Rosny-Aı̂né, J. H. 216 Huesos 193, 249
Ruff, Christopher 152, 198 Smith, Holly 154
Smith, John Maynard 265, 266
Saccopastore deposits 211 snout 20–1
sacrum 71 social behavior 23, 156, 164–5, 169,
Sadiman volcano 74 251–2
sagittal ridge 55, 81, 90, 91, 92, 95 social biology 78
Saint-Césaire 214 social skills 127
Sambungmacan braincase 223 Socrates 171
Sangiran site 179–80, 181, 223 solstice 40, 41
Sarah (chimpanzee) 241 South Africa 94, 173, 224, 236
Savage-Rumbaugh, Sue 101 South Pole 40
savanna 43–4, 76, 87, 91, 143, 173 Spain 191, 211, 214; see also individual sites
Schaik, Carel van 166 spears 198, 199, 213
Schmid, Peter 76 specializations 35–6, 94–5, 180, 266
Schmitt, Eric-Emmanuel 255 species: emergence 36; evolution 9–10;
Schmitz, Ralph 233 genes 49–50, 216–17;
Schöningen deposit 198 groupings 17–18; Linnaean
Schrenk, Friedmann 94, 98 classification 22–3;
Schutz, Adolph 160–1 paleoanthropology 56
seafaring 190 speech organs 243–4, 254
seasons 40, 41 speech production 244–7
sedimentary deposits 58, 59 speleothems 60, 194
seeds 139, 144 Spielberg, Steven 233
selection: artificial 11, 35, 238; group 249, spine 73, 74, 79
251; random 12; sexual 160–2; see also Spoor, Fred 214
natural selection Steinheim 191, 208, 248
Semendeferi, Katerina 126 Sterkfontein site 62, 71, 104
Senut, Brigitte 76 Stern, Jack 76
sex/love 163–4 stomach size 142–4
sexual dimorphism 56; Stone, Anne 233
australopithecines 84, 85, 165–6, 167, stone tools 100–2, 132, 144, 175;
168, 173; chimpanzees 159–60; China 181; Fuente Nueva-3 186; Gran
gorillas 159; Homo 102, 161; Dolina 185; Hadar 98;
Paranthropus 173; La Sima de los Neanderthals 205–6; see also
Huesos 195–6 Aurignacian; Chatelperronian;
sexual reproduction 265, 266 Mousterian
sexuality 23, 155–6 Stoneking, Mark 230, 233, 237
Seyfarth, Robert 240 Story of Life 15, 109, 256, 262, 265
Shanidar 207 Strassmann, Beverly 165
sharks’ fins 110 strepsirrhine primates 20–2
sheep 133 Stringer, Chris 207–8, 209, 226, 248
shoulder blades 24 sulci 124, 125, 127
shrews 118 Susman, Randall 76
siamang 24 Suwa, Gen 51, 91
Silo Cave 192 Swanscombe deposit 191, 208
La Sima de los Huesos 191, 192–6, 201, Swartkrans 91, 94, 107, 141
208, 249 Swisher, Carl 179, 223
Simpson, George Gaylord 217, 256, 264 Szathmáry, Eörs 265, 266
Sinanthropus 179 Szeletian industry 213–14

Arsuaga / Chosen Species 1405115327-6-ind Final Proof page 284 8.6.2005 5:21pm


Tabarin site 51 ‘Ubeidiya 185

Tabun cave 221 ulna 79
Tague, Robert 150 Uluzzian industry 213–14
Tanzania 74, 94, 162–3 Underhill, Peter 231
taphonomy 58 uranium series dating 60
Tarsiiformes 21 Ursidae 134
Tattersall, Ian 54–5
Taung Child 61–2, 141, 146, 153 Valdegoba deposit 192
technology 264–5 Vandermeersch, Bernard 219
teeth: apes 135–6; carnassial 132, 134; Vega, Gerardo 215
chimpanzees 138–9, 146; enamel 53, vegetables 131
113, 136–7, 153; herbivores 134; vegetarian diet 129–30, 142
hominids 59, 136–8, 166; Vértesszöllös deposit 191, 207, 208
mammals 19–20, 133 ; vervet monkeys (Cercopithecus
Paranthropus 139, 140, 153; aethiops) 240, 241
Pleistocene 191–2; primates 27, 134–6; Victoria, Cueva 182
size 138–41; see also canines; dentition; Vikings 37
incisors; molars Villaverde, Valentı́n 215
Teilhard de Chardin, Pierre 17 Virgil 212
temporal lobe 117, 124, 125 vision 21
temporalis muscle 81, 93, 96 vitalism 14, 16
Terra Amata 207 Viverridae 134
Teshik Tash cave 207 vocal tract 243–4, 244–6
testicles 160–1 volcanic eruptions 39, 59, 258
Therapsida 257–8 volcanic tuffs 59–60
thermoluminescence 60–1 Vrba, Elisabeth 87
Theropithecus spp. 18, 47, 47–8, 140,
182 Wainscoat, James 232
Thieme, Hartmut 198 Walker, Alan 152, 242
Thorne, Alan 222–3, 227 Wallace, Alfred Russell 10, 253–4
Tighenif deposit 190 Washoe (chimpanzee) 241
Tinbergen, Niko 156 Weidenreich, Franz 179, 181, 227
Tobias, Phillip 104, 242 Wernicke’s area 117, 241–2
toes 24, 111 whales 118, 120
tongue 245, 247 Wheeler, Peter 78, 142–4
tool-making 62, 64, 100–2, 175; White, Tim 51, 53, 56, 104, 137, 165, 166
bipedalism 77; flakes 212, 213–14; see Whiten, Andrew 170
also stone tools Williams, Jennifer 170
tool-using 65, 98–100 Wilson, Allan 230, 237
Torres, Trino 192–3 Wolpoff, Milford 227
Tossal de la Font de Vilafamés 191 Wood, Bernard 102, 104
Toth, Nick 101
Trevathan, Wenda 147, 150 Xenophon 155
Trinchera del Ferrocarril deposits 196
Trinkaus, Erik 198 Y-chromosome 229, 231–2
Trueba, Fernando 5, 264 Yunxian site 224
Truffaut, François 264
tsunami 39 Zafarraya deposit 215
Turkana, Lake 53, 105, 172 Zhoukoudian cave 181–2, 207, 209
Turkana Boy 105, 107, 146, 152–4, 200, Zinjanthropus boisei 89
242 Zuttiyeh cave 221
Tuttle, Russell 76 zygomatic bones 93, 96, 188, 203